WO2007009457A2

WO2007009457A2 - Identification of protein ligands modifying a cellular response

Info

Publication number: WO2007009457A2
Application number: PCT/DK2006/000410
Authority: WO
Inventors: Ole Thastrup; Grith Hagel; Phaedria Marie St. Hilaire; Jens Chr. Norrild
Original assignee: 2Curex Aps
Priority date: 2005-07-18
Filing date: 2006-07-12
Publication date: 2007-01-25
Also published as: WO2007009457A3

Abstract

The present invention relates to a method for identifying protein ligands capable of modulating a cellular response. The method furthermore enables identification of the protein(s) binding to the identified ligand. The method may in particular be useful in drug screening processes and allows screening for biological activity and for drug targets in a high through-put screening format.

Description

Identification of protein ligands modifying a cellular response

All patent and non-patent references cited in the application are hereby incorporated by reference in their entirety.

Field of invention

The present invention relates to a method for identifying protein ligands capable of modulating a cellular response. The method furthermore enables identification of the protein(s) binding to the identified ligand.

The method may in particular be useful in drug screening processes and allows screening for biological activity and for drug targets in a high through-put screening format.

Background of invention

Combinatorial synthesis of peptide as well as small-molecule libraries has proven very useful as a method for generating vast numbers of highly diverse compounds (see for example Comprehensive Survey of Combinatorial Library Synthesis: 2002 Roland E. DoIIe J. Comb. Chem., 2003, pp 693 - 753). To fully exploit this high ca- pacity of combinatorial chemistry to produce huge numbers of compounds several technologies have been developed that allow screening directly on the solid support (M.Meldal, 1994, METHODS: A companion to methods of enzymology 6:417-424). In the field of drug discovery such methods have been successfully applied for example to the identification of enzyme modulators. The library can be synthesized on resin beads that each carry one specific compound, and these "one-bead-one compound" libraries are then screened against the purified biological component of interest (e.g. cellular proteins or peptides). Alternatively, libraries may be synthesized on solid supports wherein a specific area of the solid support carry one specific compound resulting in solid supports with multiple compounds attached. Before ad- vancing active compounds identified though such procedures further in the drug discovery process, the compound will have to be re-synthesized and tested for efficacy in a cell-based or in-vivo test system.

Novel ways to screen combinatorial libraries in a physiologically more relevant way are assumed to greatly accelerate the drug discovery process, and show importance in areas like chemo-genomics and chemo-proteomics. Screening of combinatorial libraries in intact cells have been done by capturing mammalian or yeast cells together with a limited number of resin-beads in a "nanodroplet" (Borchart et al. Chem Biol 1997 4:961). Compounds immobilized on the resin are released through disrup- tion of a photo-cleavable linker and the compound-associated effects on the intact cells are monitored.

In an alternative method, the compounds are released through acidolysis of resin- beads carrying the library members spread out on a lawn of mammalian cells, and the spatial localization of a cellular response is monitored and beads in that region is isolated, and the remaining compound is structure elucidated (Jayawickreme et al, 1998, Combinatorial peptide Library Protocols, Ed. Shmuel Cabilly, Humana Press, p. 107-128).

WO03/038431 describes methods for screening combinatorial bead libraries by capturing cells from body fluids. Beads comprising a compound enabling cells to adhere to said bead may be selected.

US2003/0059764 describes multiplexed cell analysis systems using non-positional or positional arrays of coded carriers.

Methods involving identification of a ligand and its protein target have been described. WO00/63694 describes a method for identifying bioactive compounds by screening a library with one proteome, and subsequently identifying proteins associ- ated with components of said library. WO2004/062553 describes a method of identifying ligand-protein pairs by screening a library with one or more proteomes and identification of ligand-protein pairs. However, these methods do not provide information regarding modulation of cellular responses. Thus in the context of screening for drugs, the identified ligands must be resynthesised and tested for their ability to modulated desired cellular responses in vitro and in vivo. Summary of invention

It is of great importance to provide new and highly efficient methods for identification of new drug candidates. In particular, it is of significance to provide screening methods which select the most promising drug candidates based on relevant information regarding their cellular effects and targets. Methods wherein a very large number of candidate compounds may be tested for a specific effect on a cell allow- ing easy identification of target proteins within a relatively short period of time are therefore highly desirable.

It is therefore an object of the present invention to provide very efficient procedures for testing or discovering the influence of protein ligands from a library of test com- pounds on a physiological process in a cell and straight forward identification of protein targets of the ligand. In particular, the method provides means for testing very large numbers of different test compounds for one or more physiological effects as well as for their affinity to proteins within a rather short time period. This may be achieved by attaching living cells to resin beads coupled to a test compound. The test compounds, which optionally may be partially released from the resin bead, may thus influence physiological processes in said cells. Said influence(s) may be detected and beads containing cells displaying the desired influence(s) may be selected. Selected resin beads may then be incubated with one or more proteomes and resin beads comprising a compound associating with members of said pro- teome may be selected. Once selected, the compounds coupled to the selected beads may be identified. In addition, associated proteins may be identified. These methods may for example be very useful in connection with screening for new drugs, testing of substances for toxicity, identifying drug targets for known or novel drugs.

Thus it is a first object of the present invention to provide a method of identifying a protein ligand modifying at least one cellular response, said method comprising the steps of: (a) Providing multiple solid supports capable of supporting growth of cells, wherein each solid support is covalently linked to one member of a library of test compounds and wherein at least two solid supports comprise different library members; and

(b) Attaching cells onto said solid support, wherein said cells comprise reporter sys- tem(s) for each cellular response, wherein said reporter system(s) generate detectable outputs or can be linked to a detectable output; and

(c) Screening said solid supports for solid supports comprising cells meeting at least one predetermined selection criterion, wherein said selection criterion is linked directly or indirectly to said detectable output; and (d) Selecting solid supports comprising cells meeting said at least one selection criterion, thereby obtaining selected solid supports; and

(e) Releasing cells from the selected solid supports by cleaving the cleavable linker; and

(f) Providing one or more protein mixture(s); and (g) Incubating the selected solid supports with the one or more protein mixtures; and (h) Detecting ligand-protein binding pairs;

(i) Isolating solid supports comprising ligand-protein binding pairs; and (j) Identifying said the library member of the igand-protein binding pair, thereby identifying a protein ligand modifying said at least one cellular response,

wherein the steps of the method may be performed in any suitable order.

In general, step (a) will be performed first and step (j) will be performed last. Furthermore, steps (b), (c), (d) and (e) are in general performed one after the other in the indicated order. Similarly, steps (f), (g), (h) and (i) are in general performed one after the other in the indicated order.

Thus, in one embodiment the steps of the method are performed in the order (a), (b), (C), (d), (e), (f), (g), (h), (i) and G)

In another embodiment the steps of the method are performed in the order (a), (f), (g), (h), (i), (b), (C), (d), (e) and (j).

The method of the invention may comprise additional steps. Thus, the method may furthermore comprise the step of identifying the protein member of the ligand-protein binding pair. This step is also referred to as (k) herein. Step (k) is preferably performed immediately following step (i) or step (J)-

Preferred methods for performing the individual steps of the method are described in detail herein below in the section "Detailed description of the invention".

The cellular responses that may be modified by compounds of the invention are described in the sections "Cell surface molecule", "Cellular molecules" and "Cellular response".

Useful solid supports as provided in step (a) are described in the section "Resin beads", suitable adhesion compounds as provided in step (a) are described in the section "Cell attachment to resin beads and cell cultivation", cleavable linkers linking the adhesion compound to the solid supports provided in step (a) are described in section "Release of library compounds or of adhesion compound". Useful libraries of test compounds provided in step (a) as well as methods for preparing them are described in the section "Libray of test compounds".

Cells as provided in step (b) as well as methods for attaching cells to solid supports as outlined in step (b) are described in the section "Cells".

Suitable reporter systems comprised within the cells provided in step (b) are decsribed in the section "Reporter systems" and examples of detectable outputs generated by or linked to the reporter systems as outlined in step (b) are described in the section "Detectable output".

Suitable methods for screening and selection of solid supports meeting at least one selection criterion as outlined in steps (c) and (d) are described in the section "Selection related to cellular response".

Methods of releasing cells from solid supports by cleaving the cleavable linker as outlined in step (e) are described in the section "Release of library compounds or of adhesion compound". Useful protein mixtures as provided in step (f) are described in the section "Pro- teome".

Suitable methods for detecting ligand-protein binding pairs as outlined in step (h) and for isolating solid supports comprising ligand-protein binding pairs as outlined in step (i) are described in the section "Detecting and isolating ligand-protein binding pairs".

Useful methods for identification of the library member as outlined in step 0) and the protein as outlined in step (k) are described in the section "Identification of compound".

Description of Drawings

Figure 1 illustrates a schematic overiew of one embodiment of the invention. Primary cells are obtained from a patient. The cells may be grown on resin beads and used functional screens or for in or ex vivo analysis or for preparation of lysates for proteome screening.

Figure 2 illustrates resin beads with primary rat myocytes.

Figure 3 illustrates results of a membrane integrity assay, wherein negative control beads are mixed with positive control beads.

Figure 4 illustrates Electrospray Mass spectra.

Definitions

Naturally occurring amino acids are named herein using either their 1 -letter or 3- letter code according to the recommendations from IUPAC, see for example http://www. chem.qmw.ac.uk/iupac. If nothing else is specified amino acids may be of D or L-form. In the description (but not in the sequence listing) 3-letter codes starting with a capital letter indicate amino acids of L-form, whereas 3-letter codes in small letters indicate amino acids of D-form. The term "a" as used herein, can mean one or more, depending on the context in which it is used.

In the present context, the term "green fluorescent protein" or (GFP) is intended to indicate a protein which, when expressed by a cell, emits fluorescence upon exposure to light of the correct excitation wavelength (cf. [(Chalfie et al.1994)]). "GFP" as used herein means any protein or fragment thereof capable of fluorescing when excited with appropriate radiation. This includes fluorescent proteins that are either naturally occuring or engineered and proteins that have been modified to be fluorescent. Naturally occuring fluorescent proteins have been isolated from the jellyfish, Aequorea vistoria, the sea pansy, Renilla reniformis, Phialidium gregarium and Dis- cosoma coral (W.W. Ward et al. (1982) Photochem. Photobiol, 35:803-808; Levine et al. (1982) Biochem. Physiol., 72B:77-85; Fradkov et al. (2000), FEBS Lett. 479:127-130). GFPs have also been engineered to emit different colors and to fluoresce more intensely in mammalian organisms (U.S. Pat. Np. 5,625,048; WO 97/28261 ; WO 96/23810; EP0851874; US6,172,188; WO01/98338). A variety of Aequorea-related fluorescent proteins have been engineered to have different excitation and emission spectra by modifying the naturally occuring amino acid sequence (D.C. Prasher et al. (1992) Gene 111 :229-233; Heim et al. (1994)

Proc. Natl. Acad. Sci. USA 91 : 12501-12504; US. Pat. No. 5,625,048; WO 96/23810 and PCT/US97/14593).

The term "living cell" is used to indicate a cell which is considered living according to standard criteria for that particular type of cell such as maintenance of normal membrane potential, cell membrane integrity and energy metabolism

The term "determining the fluorescence" is used to describe the process used to monitor a change in fluorescence properties.

The term "bioluminescence" is used to describe a process where light is produced through a chemical reaction that natively is occuring in a biological system. For the reaction to occur at least two chemicals are required: the one that produces the light (called "luciferin") and the other (called "luciferase") that catalyzes the reaction. Sometimes the luciferin and luciferase are brought together in one single unit (called "photoprotein" an example of the last group is aequorin.

The term "FRET" is used to describe the occurrence of Fluorescence resonance energy transfer between a fluorophore donor and an acceptor chromophore. It is a distance-dependent interaction between the electronic excited states of two fluoro- phores in which excitation is transferred from a donor fluorophore to an acceptor chromophore without emission of a photon. The efficiency of FRET is dependent on the inverse sixth power of the intermolecular separation, making it useful over dis- tances comparable with the dimensions of biological macromolecules. Thus, FRET is an important technique for investigating interactions between cellular molecules for example complex formation.

The term "BRET" is used to describe a process that is related to FRET, but differs from FRET in that donor is a bioluminescent protein like luciferase that generates its own luminescence emission in the presence of a substrate, and that can pass the energy to an acceptor fluorophore. For either BRET or FRET to work, the donor's emission spectrum must overlap the acceptor's absorption spectrum, their transition dipoles must be in an appropriate orientation, and the donor and acceptor must be in close proximity (usually within 30-80 A of each other, depending on the degree of spectral overlap).

The term "Scintillation Proximity Assay" is used to describe an assay determining the distance between two compounds, wherein one compound (bound to a bead) will emit light when radiation from an isotope occurs in close proximity and the other compound is containing a radioactive isotope.

A "Protein ligand" is a compound capable of binding a protein with a preferred affinity ranging from micro- to picomolar and even more preferred affinity of nano- to picomolar.

The term "mammalian cell" is intended to indicate any cell of mammalian origin. The cell may be an established cell line, many of which are available from The American Type Culture Collection (ATCC, Virginia, USA) or a primary cell with a limited life span derived from a mammalian tissue, including tissues derived from a transgenic animal, or a newly established immortal cell line derived from a mammalian tissue including transgenic tissues, or a hybrid cell or cell line derived by fusing different celltypes of mammalian origin e.g. hybridoma cell lines. The cells may optionally express one or more non-native gene products, e.g. receptors.

The phrase "fluorescence properties" means absorption properties, such as wavelength and extension, or spectral properties of the emitted light, such as wavelength, fluorescence lifetime, intensity or polarisation, or the intracellular localisation of the fluorophore. It may thus be localised to a specific cellular component (e.g. organelle, membrane, cytoskeleton, molecular structure) or it may be evenly distributed throughout the cell or parts of the cell.

The term "fixed cells" is meant to cover cells treated with a cytological fixative such as glutaraldehyde, methanol, acetone or formaldehyde, treatments which serve to chemically cross-link and/or stabilize soluble and insoluble proteins within the structure of the cell or to dehydrate cells. Once in this state, such proteins cannot be lost from the structure of the now-dead cell.

The term "cell line" is meant to cover a group of cells, wherein the cells of that group are essentially genetically indistinguishable from each other. The cells of a cell line are thus all progeny of the same cell.

The term "comprising" should be understood in an inclusive manner. Hence, by way of example, a composition comprising compound X, may comprise compound X and optionally additional compounds.

The term "multiple" should be understood as "at least two".

The term "library of test compounds" should be understood as a collection of test compounds comprising at least 2 different test compounds.

The term "small organic molecules or compounds" refers herein to non-oligomeric, carbon containing compounds producible by chemical synthesis and generally having a size of less than 600 mass units. The term "one bead-one compound library" refers to libraries immobilised on resin beads, wherein each individual resin bead does not comprise more than one library member in one or multiple copies. In a particular form of such libraries each member is represented by multiple fragments of said member obtained by ladder synthesis encoding.

The term "one bead-two compound library" refers to libraries immobilised on resin beads, wherein each individual resin bead does not comprise more than one library member in one or multiple copies and wherein each individual resin bead in addition to said library member also comprises an adhesion compound. All beads may comprise identical adhesion compounds.

Detailed description of the invention

Library of test compounds

In the present invention, libraries of compounds are used to screen for compounds having a desired physiological influence on a living cell. As used herein, the term "library" means a collection of molecular entities or test compounds, herein also designated "library members" obtained after a series of chemical transformations.

In preferred embodiments of the present invention the library is a combinatorial library. Non-limiting examples of combinatorial libraries that may be used with the present invention and methods of producing such libraries are given in: Comprehensive Survey of Combinatorial Library Synthesis: 1998 Roland E. DoIIe and Kingsley H. Nelson, Jr. J. Comb. Chem., 1999, pp 235 - 282; Comprehensive Survey of Combinatorial Library Synthesis: 1999 Roland E. DoIIe J. Comb. Chem., 2000, pp 383 - 433; Comprehensive Survey of Combinatorial Library Synthesis: 2000 Roland E. DoIIe J. Comb. Chem. ,2001 , pp 477 - 517; Comprehensive Survey of Combinatorial Library Synthesis: 2001 Roland E. DoIIe J. Comb. Chem.,2002, pp 369 - 418 and Comprehensive Survey of Combinatorial Library Synthesis: 2002 Roland E. DoIIe J. Comb. Chem.,2003, pp 693 - 753. The skilled person will appreciate that these protocols may easily be adapted to specific need of a particular embodiment of the present invention. In one embodiment, the test compounds can be natural oligomers (oligomers of building blocks occurring in nature) such as peptides, glycopeptides, lipopeptides, nucleic acids (DNA or RNA), or oligosaccharides. By way of example, a natural oli- gomer may be any peptide consisting of naturally occurring amino acid, even if said peptide comprises a sequence not present in nature. The libraries may comprise different natural oligomers or the libraries may comprise only one kind of natural oligomer, for example the library may be a peptide library. In another embodiment, they can be unnatural oligomers. Unnatural oligomers are oligomers comprising one or more building blocks not occurring in nature. Thus unnatural oligomers may consist of building blocks not occurring in nature or they may comprise a mixture of building blocks occurring in nature and building blocks not occurring in nature, such as chemically modified peptides, glycopeptides, nucleic acids (DNA or RNA), or, oligosaccharides, and the like. Said chemical modification may for example be the use of unnatural building blocks connected by the natural bond linking the units (for example, a peptide amide linkage), the use of natural building blocks with modified linking units (for example, oligoureas as discussed in Boeijen et al, 2001 , J. Org. Chem., 66: 8454-8462; oligosulfonamides as discussed in Monnee et al, 2000, Tetrahedron Lett., 41 : 7991-95), or combinations of these (for example, statine amides as discussed in DoIIe et al, 2000, J. Comb. Chem., 2: 716-31.). Preferred unnatural oligomers include oligomers comprising unnatural building blocks connected to each other by a naturally occurring bond linking. Said oligomers may thus comprise a mixture of naturally occurring and unnatural building blocks linked to each other by naturally occurring bonds. By way of example, the oligomer may comprise naturally occurring amino acids and unnatural building blocks linked by peptide bonds f.x. PNA or LNA. Thus, in one embodiment of the invention preferred oligomers comprise modified amino acids or amino acid mimics). Other preferred unnatural oligomers include, for example oligoureas, poly azatides, aromatic C-C linked oligomers and aromatic C-N linked oligomers. Still other preferred oligomers comprise a mixture of natural and unnatural building blocks and natural and unnatural linking bonds. For example, the unnatural oligomer may be any of the oligomers mentioned in recent reviews see: Graven et al., 2001 , J. Comb. Chem., 3: 441-52; St. Hilaire et al., 2000, Angew. Chem. Int. Ed. Engl., 39: 1162-79; James, 2001 , Curr. Opin. Pharmacol., 1 : 540-6; Marcaurelle et al., 2002, Curr. Opin. Chem. Biol., 6: 289-96; Breinbauer et al., 2002, Angew. Chem. Int. Ed. Engl., 41 : 2879-90. The libraries of the invention may also comprise cyclic oligomers, for example cyclic natural oligomers, such as cyclic peptides or cyclic unnatural oligomers. In certain embodiments of the invention, libraries of cyclic oligomers may be advantegous to use due to the rigid structure. This may result in higher selectively and affinity.

In yet another embodiment, the molecular entities may comprise non-oligomeric molecules such as peptidomimetics or other small organic molecules. Peptidomi- metics are compounds that mimic the action of a peptidic messenger, such as bi- cyclic thiazolidine lactam peptidomimetics of L-proplyl-L-leucyl-glycinamide (Khalil et al, 1999, J. Med. Chem., 42: 2977-87). In a preferred embodiment of the invention, the library comprises or even more preferably consists of small organic molecules. Small organic molecules are non-oligomeric compounds of less than about 600 mass units containing any of a variety of possible functional groups and are the product of chemical synthesis, or isolated from nature, or isolated from nature and then chemically modified, and include, for example, urea-based kinase inhibitors (Smith et al., 2001 , Bioorg. Med. Chem. Lett., 11 : 2775-78). Small organic compounds may for example be selected from the group consisting of alcohols, ethers, carboxylic acids, aryloxy, acyloxy, thiol, alkylthio, arylthio, heteroarylthio, sulphonyl, sulphoxy, amino, alkylamino, dialkylamino, acylamino, diacylamino, alkoxycarbonyl- amino, amides, alkyl, branched alkyl, aryl, heteroaryl, nitro, cyano, halogeno, sily- loxy, keto, heterocycles, fused ring systems, fused heterocycles and mixtures thereof, wherein each of the aforementioned may be substituted independently on each position with one or more groups selected from the group consisting of -H, - OH, -SH, halogen, carboxyl, carbonyl, alkoxy, aryloxy, acyloxy, alkylthio, arylthio, heteroarylthio, sulphonyl, sulphoxy, amino, alkylamino, dialkylamino, acylamino, diacylamino, alkoxycarbonylamino, amides, alkyl, aryl, heteroaryl, nitro, cyano, halogeno, silyloxy, keto, heterocycles, fused ring systems, and fused heterocycles.

In a preferred embodiment the small organic molecule libraries are prepared starting from one or more basic structures. Said basic structures may for example be selected from the group consisting of alkoxy, aryloxy, acyloxy, thiol, alkylthio, arylthio, heteroarylthio, alkylamino, dialkylamino, acylamino, diacylamino, alkoxyacylamino, dialkoxyacylamino, amides, alkyl, branched alkyl, aryl, heteroaryl, keto, heterocycles, fused ring systems, fused heterocycles and mixtures thereof, wherein each of the aforementioned may be substituted with one or more groups selected from the group consisting of -H, -OH, -SH, halogen, carboxyl, carbonyl, alkoxy, aryloxy, acy- loxy, alkylthio, arylthio, heteroarylthio, sulphonyl, sulphoxy, amino, alkylamino, dial- kylamino, acylamino, diacylamino, alkoxyacylamino, dialkoxyacylamino, amides, alkyl, aryl, heteroaryl, nitro, cyano, halogeno, silyloxy, keto, heterocycles, fused ring systems, and fused heterocycles. In one embodiment, the basic structure may for example be a halide containing aromatic or heteroaromatic carboxylic acid or acid halide. Accordingly, small organic compound libraries may for example comprise compounds comprising one or more aromatic or hetero aromatic ring system(s), wherein the heteroatoms is preferably O, S and/or N, one or more non-aromatic ring systems, which may or may not comprise heteroatoms or various substituted alkyls. Said aromatic or non-aromatic ringsystems may be fused. Each of the aforementioned may be substituted with various substituents e.g. Halide(s) (F, Cl), CH3-, CF3-, Methoxy-, thiomethyl-, aldehyde(s)-, carboxylic acids-, esters-, nitrogroups, - H, -OH, -SH, carbonyl, alkoxy, aryloxy, acyloxy, alkylthio, arylthio, heteroarylthio, sulphonyl, sulphoxy, amino, alkylamino, dialkylamino, acylamino, diacylamino, alkoxyacylamino, dialkoxyacylamino, amides, alkyl, aryl, heteroaryl, cyano, halogeno, silyloxy or keto. The small organic compound libraries according to the invention may also comprise mixtures of any of the above mentioned compounds.

Non-limiting examples of small organic molecule libraries that may be used with the present invention and methods of producing them may for example be found in the reviews Thompson et al., 1996, Chem. Rev., 96: 555-600; Al-Obeidi et al., 1998, MoI. Biotechnol., 9: 205-23; Nefzi et al., 2001 , Biopolymers, 60: 212-9; DoIIe, 2002, J. Comb. Chem., 4: 369-418.

The libraries according to the invention may comprise at least 20, such as at least 100, for example at least 1000, such as at least 10,000, for example at least 100,000, such as at least 1 ,000,000 different test compounds. Preferably, the libraries comprises in the range of 20 to 10,000,000, more preferably 50 to 7,000,000, even more preferably 100 to 5,000,000, yet more preferably 250 to 2,000,000 different compounds. In a very preferred embodiment of the present invention the libraries comprises in the range of 1000 to 20,000, such as in the range of 20,000 to 200,000 different test compounds. In preferred embodiments of the invention the library comprises in the range of 10,000 to 1 ,000,000 different test compounds. Preferably, the libraries to be used with the present invention are immobilised on resin beads. Said resin beads may be any of the beads described herein below. At least 2, preferably at least 20, more preferably at least 100, even more preferably at least 1000, yet more preferably at least 10,000, for example at least 100,000, such as at least 1 ,000,000 resin beads comprising different library members, i.e. different test compounds may be used with the methods according to the invention. Preferably, the in the range of 20 to 10,000,000, more preferably 100 to 7,000,000, even more preferably 1000 to 5,000,000, yet more preferably 5000 to 2,000,000, even more preferably 10,000 to 1 ,000,000 resin beads comprising different library mem- bers, are used with the methods according to the invention.

In one very preferred embodiment of the invention, each resin bead does not comprise more than one library member in one or more copies, i.e. each resin bead only comprises one kind of test compound, however said test compound may be present on the resin bead in multiple copies. Such libraries may also be designated one- bead-one-compound libraries. Preferably, each resin beads comprises sufficient copies of said library member in order to exert the desired influence of cells attached to said resin bead and in order to analyse the chemical structure of the compound. Such libraries may be prepared by different methods, for example by a split/mix method or by coupling individually a specific compound to a bead. One-bead-one compound libraries offer the advantage that once a resin bead has been selected according to the methods described herein, the desired compound may easily be identified (see useful methods herein below).

The libraries may in one preferred embodiment be synthesized directly on resin beads using a split/mix method (vide infra), which gives, rise to one-bead-one- compound libraries. Split/mix methods in general comprise the steps of:

1. Providing several pools of resin beads

2. Performing one or more different chemical synthesis steps on each pool of resin beads,

3. Splitting said pools to obtain fractions

4. Mixing fractions from different pools, thereby obtaining new pools

5. Optionally repeating steps 1 to 4

Alternatively steps 3 and 4 may be as follows: 3. Mixing all pools of resin beads, thereby obtaining a mixed pool

4. Splitting the mixed pool of resin beads into reaction containers thereby obtaining new pools.

One-bead-one-compound libraries may for example be prepared as described in M. Meldal, Multiple column synthesis of quenched solid-phase bound fluorogenic substrates for characterization of endoprotease specificity in Methods: A Companion to Methods in Enzymology 6:417-424, 1994 or in M. Meldal, The One-bead Two- Compound Assay for Solid Phase Screening of Combinatorial Libraries in Biopoly- mers, Peptide Science 66:93-100, 2002; or in Combinatorial peptide library protocols, Ed. by Shmuel Cabilly, Humana Press, 1998, p. 1 -24 and 51 to 82. In another embodiment of the invention the library may be a one-bead-two- compounds library. Each individual resin bead of such a library comprises only one library member in one or more copies. In addition each individual resin bead comprises a second compound, such as a cell adhesion compound. The cell adhesion compound could for example be any of the cell adhesion compounds mentioned herein below. It is comprised within the invention that several library resin beads, such as all library resin beads comprise identical adhesion compound(s) in one or more copies. One-bead-two-compound libraries may for example be prepared by a method involving the steps of:

1. Providing resin beads comprising a plurality of reactive groups

2. Reacting said reactive groups with two chemical moeities comprising different and preferably orthogonal protective groups 3. Deprotecting a subset of the reactive groups by removal of one kind of protective groups, preferably selective removal of one kind of protective group,

4. Attaching or synthezising a split/mix library of test compounds to the deprotected reactive group

5. Deprotecting the remaining reactive groups by removal the other kind of protec- tive group

6. Attaching the second compound to the deprotected reactive groups

The method may also be performed by first attaching the second compound and then synthesising the library. Accordingly, the steps of the method may be per- formed in the following order: 1 , 2, 3, 6, 5 and 4. The library of test compounds may be first synthesized and then attached to the resin beads or it may be synthesized directly onto the resin bead. Similarly, the second compound may be first synthesized and then attached to the resin beads or it may be synthesized directly onto the resin bead.

Preferred resin beads are described in the section "resin beads" herein below. The reactive group may be any suitable reactive group, preferably however, the reactive group is either a hydroxy! group, a thiol or a primary amino group. The reactive may also preferably be an azido or a secondary amino group. The protective group may be any suitable protective group known to the person skilled in the art, such as acid labile, alkaline labile or photolabile protective groups; preferably the protective group is selected from the group consisting of Fmoc, Boc, Alloc and N₃. It is preferred that the different protective groups may be removed by different treatment, for example that if one protective group is acid labile, then the other is not acid labile, but instead for example alkaline labile or photo labile. In a preferred embodiment one protective group is Fmoc and the other protective group is Alloc or N₃. Step 4 may for example be performed by a split/mix method as described herein above, thereby generating a one-bead-one-compound library. The second compound is preferably a cell adhesion compound

In one embodiment the library may be linked to the resin bead via a linker, which may be a cleavable linker. This may for example be achieved by synthesizing the linker directly on resin beads or coupling the linker to the resin beads and subsequently coupling or synthesizing the library onto the resin beads. Thus, before cou- pling of the library the linker preferably comprises a protective group as described herein above. The cleavable linker may be any of the cleavable linkers described herein below. If the resin beads are coupled to an adhesion compound via a cleavable linker it is preferred that the cleavable linker linking the library is different to the cleavable linker linking the adhesion compound. It is in particularly preferred that the linkers are not cleavable by the same mechanism. Thereby, the library may be specifically released from the resin beads, without release of adhesion compounds. In yet another embodiment of the invention the library may be a mixed compound library, wherein each individual resin bead comprises a plurality of library members. Selection of an appropriate library is dependent upon the specific embodiment of the invention. For example, a totally random library designed to contain interesting and greatly diverse compounds may be used with the invention. An advantage of this approach is that the outcome of the screening is not prejudiced in any specific manner. Since the invention permits screening of millions of diverse compounds, for example, immobilized on resin beads, a large number, for example in the range of 3 to 5 million, of random molecules can be used in the ligand library.

Alternatively, a smaller, targeted library (hundreds to thousands of compounds) can be used, for example, starting with a known compound or compounds, and providing numerous variations of these known compounds for targeted screening. For exam- pie, in embodiments of the invention wherein compounds modulating the activity of a specific cell surface molecule, a compound known to modulate said specific cell surface molecule may be used as starting compound for the preparation of a targeted library. Alternatively, a smaller targeted library of compounds mimicking a compound known to modulate the activity of said cell surface molecule may be pre- pared, for example using computer aided modelling followed by chemical synthesis. The smaller, targeted library can also comprise random molecules. Examples of libraries and methods of preparing such libraries, which may useful in embodiments of the invention, wherein the cellular response is mediated through a G-protein coupled receptor are described in C. Haskell-Luevano, A. Rosenquist, A. Souers, K. C. Khong, J. A. Ellman, and R. D. Cone, 1999, J.Med.Chem. 42:4380-4387. Compounds that activate the mouse melanocortin-1 receptor identified by screening a small molecule library based upon the b-turn. J.Med.Chem. 42:4380-4387, 1999; A. J. Souers, A. A. Virgilio, A. Rosenquist, W. Fenuik, and J. A. Ellman. Identification of a potent heterocyclic ligand to somatostatin receptor subtype 5 by the synthesis and screening of b-turn mimetic libraries. J.Am.Chem.Soc. 121 (9):1817-1825, 1999; J. Bondebjerg, Z. Xiang, R. M. Bauzo, C. Haskell-Luevano, and M. Meldal. A solid phase approach to mouse melanocortin receptor agonists derived from a novel thio- ether cyclized peptidomimetic scaffold. J.Am.Chem.Soc. 124:11046-11055, 2002; B. A. Harrison, G. W. Pasternak, and G. L. Verdine. 2,6-dimethyltyrosine analogues of a stereodiversified ligand library: highly potent, selective, non-peptidic m opioid receptor agonists. J.Med.Chem. 46:677-680, 2003; G. R. Marshall. Peptide interactions with G-protein coupled receptors. Peptide Science 60:246-277, 2003; P.N.Arasasingham, C.Fotsch, X.Ouyang, M.H.Norman, M.G.Kelly, K.L.Stark, B.Karbon, CHaIe, J.W.Baumgartner, M.Zambrano, J.Cheetham, N.A.Tamayo, and . Structure-Activity relationship of (1 -aryl-2-piperazinylethyl) piperazines: Antagonists for the AGRP/Melanocortin receptor binding. J.Med.Chem. 46:9-11 , 2003. Further useful libraries are described in examples 4, 5, 5a, 6a and 6b in PCT/DK2005/000348. The person skilled in the art will appreciate that other libraries may be prepared by adapting the protocols described in the aforementioned refer- ences. The library may contain a parallel array of random modifications of one or more compounds. In one embodiment, the library may be formed as a parallel array of random modifications to a known compound or compounds. The term "parallel array" is meant to cover synthesis of a library by subjecting a given compound to a known set of reactions in an isolated vessel or well. Thus, the identity of a com- pound in a given container or well is known. The array of test compounds is preferably prepared directly on resin beads using techniques known by those skilled in the art. Briefly, the resin may be portioned into a number of vessels or wells, usually less than 500 and the reagents added. There is in general no mixing step and after the appropriate washing steps, subsequent reactions are carried out by addition of additional reagents to the wells. There is no exponential increase in the number of compounds generated and that is equal to the number of vessels used. The compound can be easily identified by keeping track of the reagent added to each well.

The library may also have been prepared by using a tag to enable identification of, what chemical synthesis steps the individual resin bead has been submitted to. This may for example be done by IRORI or radiofrequency tag. Alternatively, chemical synthesis steps may be performed in parallel to preparing a polymeric tag. Identification of the tag will thus provide knowledge of the compound. Alternatively, optical encoding of individual resin beads may be obtained as described in WO2005/061094.

Attachment of a label to a compound may alter the properties of said compound. Hence, in one embodiment of the present invention, the compounds of the library are not labelled, i.e. the compounds are not connected to a detectable label, such as a fluorescent component, a nucleic acid or a nucleic acid homologue such as PNA, a dye, a probe comprising a reactive moiety or the like. In particular it is preferred that all compounds are not connected to the same detectable label.

Examples of specific libraries that may be used with the present invention include but are not limited to the following libraries: -Libraries prepared as described in examples 4, 5, 5a, 6a and 6b of PCT/DK2005/000348, which is hereby incorporated by reference in its entirety.

-Libraries prepared as described in examples 4, 5a, 5b and 5c of

PCT/DK2005/000347, which is hereby incorporated by reference in its entirety.

-Library 1 prepared as described in example 5, library 2 prepared as described in example 6, library 3 prepared as described in example 7, library 4 prepared as de- scribed in example 35, library 5 prepared as described in example 36, library 6 prepared as described in example 39 and library 7 prepared as described in example 40 of WO2004/062553.

-Libraries prepared taking advantage of intramolecular Pictet-Spengler reactions as described in WO2004/113362, which is hereby incorporated by reference in its entirety.

Resin beads

The library members of this invention are preferably bound to a solid support. Preferred solid supports to be used with the present invention are resin beads (see herein below).

The solid support may however also be a spot or region on a surface or a plated gel or a membrane. A spot or a region is a defined area on said surface, to which the library member is covalently bound. One can therefore envisage one surface comprising a plurality of spots or regions, wherein each such spot or region is covalently attached to only one library member in one or more copies. Said surface could for example be a silicium wafer, a glass surface, a plastic surface or a gel.

Plastic surfaces may for example be prepared from polystyrene, polycarbonate polypropylene, ethylene and/or teflon. Gels could be prepared from for example poly acrylamide or PEGA. In this invention however, the compounds of the library are preferably bound to a resin bead, conferring the advantage of compartmentalized "mini-reaction vessels" for attachment of cells and subsequently for sceening of proteomes.

In general more compounds may be screened and several of the steps in the procedure may be performed on one bead with sufficient material. Hence, preferably, the library is bound to resin beads. Each member of the library is a unique compound and is physically separated in space from the other compounds in the library, preferably, by immobilizing the library on resin beads, wherein each bead at the most comprises one member of the library. Depending on the mode of library synthesis, each library member may contain, in addition, fragments of the library member. Since ease and speed are important features of this process invention, it is preferred that the screening step take place on the same solid support used for synthesis of the library, and also that identification of the members of the binding pair can take place on the same support, such as on a single resin bead. Thus, preferred solid supports useful in the process invention satisfy the criteria of not only being suitable for organic synthesis, but are also suitable for screening procedures, such as "on- bead" screening as well as suitable for attachment of cells. It is furthermore preferred that the resin bead is suitable for "on-bead" identification of library members as described herein below. The resin bead may be prepared from any suitable material such as polystyrene, polyethylene polyacrylamide, controlled pore glass or PEG. The resin bead could thus for example be selected from the group consisting of Toyopearl, sepharose, sephadex, CPG, silica, POEPOP, PEGA, SPOCC, Expan- sin, TentaGel, ArgoGel, Polystyrene, Jandagel, polydimethylacrylamide resin, PoIy- acrylamide resin, kieselghur supported resins and polystyrene supported resins.

Hydrophilic supports are preferred. Examples of preferred hydrophilic resin beads includes TentaGel (commercially available from Rapp polymere, Tubingen, Germany), ArgoGel (commercially available from Argonaut Technologies Inc., San Car- los, CA), PEGA (PolyEthyleneGlycol Acrylamide copolymer; Meldal M., 1992, Tetrahedron Lett., 33: 3077-80; commercially available from VersaMatrix, Copenhagen), POEPOP ((PolyOxyEthylene-PolyOxyPropylene; Renil et al., 1996, Tetrahedron Lett., 37: 6185-88; available from Versamatrix, Copenhagen, Denmark) and SPOCC (Super Permeable Organic Combinatorial Chemistry; Rademann et al, 1999, J. Am. Chem. Soc, 121 : 5459-66; available from Versamatrix, Copenhagen, Denmark). Examples of on-bead screening attempts are described in the following references: Chen et al., 1996, Methods Enzymol., 267: 211-19; Leon et al., 1998, Bioorg. Med. Chem. Lett., 8: 2997-3002; St. Hilaire et al., 1999, J. Comb. Chem., 1 : 509-23; Smith et al., 1999, J. Comb. Chem., 1 : 326-32; Graven et al., 2001 , J. Comb. Chem. 3: 441 -52; Park et al., 2002, Lett. Peptide ScL, 8: 171 -78). TentaGel and ArgoGel are made up of polyethylene glycol chains grafted on to a polystyrene core. However, use of these supports in biological screening is limited by a size restriction, and by denaturation of certain proteins, particularly enzymes.

Preferred resin beads according to the present invention are resin beads, useful for on-bead library synthesis, screening and identification of ligand/protein. Hence, preferred resins according to the present invention are resin comprising polyethylene glycol. More preferably, the resin is PEGA, SPOCC POEPOP resin. Another preferred resin comprises a crosslinked polyacrylamide resin. PEGA, POEPOP and SPOCC resins are made primarily of polyethylene glycol and swell well in organic as well as aqueous solvents. Because they have very reduced or no non-specific binding, PEGA and SPOCC resins have been effectively used in the screening of myriad proteins including enzymes of different classes. Furthermore, these resins are available in different pore sizes and can allow large proteins to enter while retaining activity. For example, PEGA6000 resins allow proteins up to 600 kDa to enter. In a preferred embodiment, PEGA4000 or PEGA1900 resin with a molecular weight cut off of 200 and 55-90 kDa, respectively, are used for screening, however other PEGA resins with different poresizes may also be used. In principle, any hydrophilic support that is useful for compartmentalized synthesis, retains the activity of the pro- teins, and has minimal non-specific binding, may be used in this process invention.

One aspect of the invention relates to a method comprising the step of providing multiple resin beads capable of supporting growth of cells. Preferably, all resin beads provided are capable of supporting growth of cells. In one preferred embodi- ment all resin beads are similar and each is capable of supporting growth of cells, wherein the resin beads only differs by comprising different library members. In embodiments of the invention wherein the resin beads comprise a cell adhesion molecule, it is preferred that at least 10%, more preferably at least 20%, even more preferably at least 30%, yet more preferably at least 40%, even more preferably at least 50%, yet more preferably at least 60%, %, even more preferably at least 70%, yet more preferably at least 90%, even more preferably essentially all, yet more preferably all resin beads comprise the cell adhesion molecule as well as a library member.

Cells

The cells to be used with the present invention may be any useful cells available or prepared for the purpose. Preferably, the cells are selected from the group consist- ing of mammalian cells. For example the cells may be human cells. The cells may be cells capable of growing in suspension or they may be adherent cells. Adherent cells may preferably be cultivated directly on the resin beads used with the invention (see also herein below). It is preferred that the cells are adherent cells. Cells with a better adherence are preferred over cells with a poorer adherence. Cells which ad- here well to resin beads comprising an adhesion compound as described herein above are very preferred.

Cells could for example be primary cells or established cell lines. Preferred cell lines include but are not limited to those mentioned in Table 1 in PCT/DK2005/000348.

In one embodiment of the invention the cells have been genetically or otherwise modified in order to enhance their usability with the present invention. The modification may be stable or only transient or a mixture of both. For example, the cells may have been modified to contain one or more of the reporter systems described herein below. Depending on the nature of the reporter system this may be achieved by a number of different methods. For example, if the reporter system comprises a nucleic acid, said nucleic acid may be inserted into said cell by conventional recombinant techniques (see below).

In another preferred example the cell comprises a nucleic acid comprising a first nucleotide sequence encoding a cell surface molecule operably linked to a second nucleotide sequence not naturally associated therewith directing expression of said first sequence. The cell surface molecule may be any of the cell surface molecules described herein below. Such cells are in particular useful for identification of com- pounds modulating the activity of said cell surface molecule. Said nucleic acid may be introduced transiently or stably into said cells.

Useful second sequences include for example promoters active in the particular cells, for example mammalian promoters, viral promoters or synthetic promoters. A large number of useful eukaryotic promoters are known to the person skilled in the art, useful promoters are for example described in"Mechanism of Transcription" (1998) Cold Spring Harbor Symposia on Quantitative Biology Vol. LXIII; Cold Spring Harbor Laboratory Press. Such promoters may be constitutively active or they may be active only temporarily. In one example the promoter may be regulated by an external signal, for example the promoter may be inducible or repressable.

The nucleic acid may be inserted into the cells by any useful method, for example by conventional recombinant techniques, such as any of the techniques described in Sambrook et al., Molecular Cloning: A Laboratory Manual, 1989, Cold Spring Harbor Laboratory, New York, USA

In another embodiment the cells are primary cells. Primary cells are cells with a limited life span that preferably are derived from a mammalian tissue. Preferred primary cells are cells which are adherent. The mammalian tissue may for example be a human tissue, such as healthy or diseased tissue. In one embodiment the tissue is or comprises a neoplastic tissue, for example tissue removed from a cancer patient by surgery, for example from a patient suffering from melanoma, breast cancer or colon cancer. The tissue may also be hypertrophic cells, such as cardiac myocytes or it may be healthy or diseased liver tissue. Preferably said cancer patient has not been subjected to radiotherapy prior to surgery. In embodiments of the invention wherein the cells are primary cells it is preferred that the reporter system is endogenous to said primary cells.

Cell attachment to resin beads and cell cultivation

The present invention relates to methods comprising the step of attaching cells comprising a reporter system(s) to solid supports, such as resin beads. Preferably, the resin beads are linked to an adhesion compound either directly or via a cleav- able linker, more preferably the resin beads are linked to an adhesion compound via a cleavable linker. Any cell adhesion compound faciliating or promoting cell adherence known to the person skilled in the art may be used with the present invention, for example any of the adhesion compounds described by Hersel et al. Biomaterials 24, 4385-4415, 2003 and Georges & Janmey J Appl Physiol 98, 1547-1553. It is frequently an advantage if the cell adhesion compound comprises at least one positively charged moiety at neutral pH, more preferably the cell adhesion compound has an overall positivenetcharge at neutral pH. In another example the cell adhesion molecule may be or may comprise lipids. For example, the cell adhesion molecules may be lipidyl-PEG derivatives as described in Kato, K et al, BioTechniques, 2003 (35) 1014-1021.

In one preferred embodiment of the invention the cell adhesion compound comprises a peptide or a polypeptide, more preferably the cell adhesion compound con- sists of a peptide. Such peptides are herein also designated "adhesion peptides".

Said peptide preferably consists of in the range of 3 to 100, preferably in the range of 3 to 75, more preferably in the range of 3 to 50, even more preferably in the range of 3 to 30, yet more preferably in the range of 3 to 25, even more preferably in the range of 3 to 20, yet more preferably in the range of 3 to 15, such as in the range of 3 to 10, for example in the range of 3 to 8, for example in the range of 6 to 7 amino acids. In general, it is sufficient if the peptide comprises at least 3 amino acids.

It is preferred that the peptide comprises at least one amino acid selected from the group consisting of arginine and lysine, more preferably the peptide comprises at least 2 basic amino acids, such as 3 basic amino acids selected from the group consisting of Arg and Lys, even more preferably the peptide has an overall positive netcharge. In one preferred embodiment the peptide comprises the following sequence of 4 amino acids: basic-basic-lipophilic-basic. Basic amino acids may for example be selected from the group consisting of arginine and lysine, whereas the lipophilic amino acid may be selected from the group consisting of GIy, Ala, VaI, Leu, lie, Phe, Trp, Pro and Met of either D or L-form. Preferably, the peptide comprise at least 1 , preferably at least 2, more preferably at least 3, even more preferably at least 4 amino acid on the D-form, yet more preferably all amino acids are on the D-form. Preferably D-amino acids are used to enhance the metabolic stability but also L-amino acids may be used.

In one embodiment of the invention it is preferred that the peptide has low or essen- tially no fluorescent properties. It is particularly preferred that the peptide has low or essentially no fluorescent properties when attached to a solid support, such as a resin bead. By "essentially no fluorescent properties" is meant that the peptide does not emit any detectable fluorescence. This is in particularly relevant for embodiments of the invention wherein the detectable output is fluorescence (see herein below).

Preferred examples of peptides useful as cell adhesion compounds are given in table 2 of PCT/ DK2005/000348, in particular the peptides identified by SEQ ID: 21 to 23 and 26 to 35 in PCT/DK2005/000348 may be useful. Another useful adhesion compound is a peptide of the sequence RGD. Also peptides comprising any of the aforementioned peptides may be useful. For example, in order to immoblised the peptide on a resin bead it may be useful to synthesise the adhesion peptide on an amino acid immobilized on the resin bead, for example a GIy.

Peptides useful as cell adhesion compounds may be identified using any suitable method, such as the methods described in examples 1 and 1a of PCT/DK2005/000348.

The adhesion compound is preferably linked to the resin bead via a cleavable linker. This may for example be achieved by synthesizing the linker directly on resin beads or coupling the linker to the resin beads and subsequently coupling or synthesizing the adhesion compound onto the resin beads. Thus, before coupling of the adhesion compound the linker preferably comprises a protective group, such as Fmoc, N₃ or Alloc. In one embodiment Alloc is the preferred protective group. The adhesion compound may be synthesized onto the linker by any standard techniques, such as standard Fmoc technology as described in B. Blankemeyer-Menge, M. Nimtz, and R. Frank, An Efficient method for ancoring Fmoc-amino acids to hydroxyl- function- alised solid supports. Tetrahedron Lett. 31 :1701 -1704, 1990. Sidechains may be protected with acid labile protecting groups such as t-Bu, Trt, Pmc, Boc etc. The protected peptide may for example be cleaved off the resin using alkaline conditions or hydrazine and the structure may be determined e.g. by on bead Edman Degrad- tion. An non-limiting example of a method for synthesizing an adhesion peptide is given in example 5a, "Synthesis of adhesion peptide" in PCT/2005/000348 and in example 1 herein. It is preferred that different protecting groups are used for synthe- sis of the adhesion peptide or for library synthesis. The cleavable linker may be any of the cleavable linkers described herein below. If the resin beads are coupled to the library via a cleavable linker it is preferred that the cleavable linker linking the adhesion compound is differentially cleavable. Cells may preferably be at least partially or even more preferably essentially fully released from the resin bead by cleavage of the cleavable linker linking the adhesion compound to the resin bead. Following release of cells from the resin bead, the resin beads may be used for further screening. It is preferred within the present invention that cells are released from the resin bead prior to incubation with one or more proteomes, preferably cells are released by cleavage of the cleavable linker linking the adhesion compound to the resin bead (see details regarding release herein below).

The cells may be cultivated directly on the resin beads. In general, a method of cultivating cells on resin beads may comprise the steps of 1. Providing resin beads capable of supporting growth of cells 2. Seeding cells onto said resin bead

3. Incubating said resin beads comprising said cells in a cell culture medium under cell cultivation conditions

4. Optionally allowing said cells to divide on said resin bead Thereby cultivating cells on resin beads

The cells may adhere actively to the resin beads and will then generally be referred to as adherent cells.

Cells cultivation conditions depends on the specific cells. For a large number of mammalian cells, such conditions comprise high humidity, preferably close to 100%, approximately 5% CO2 and around 37⁰C. It is often desirable to keep the resin beads immersed in a suitable cultivation medium and frequently it is also desirable that the resin beads can be circulated within said medium, for example by stirring, gentle rocking or rotation. Said stirring or rotation may be continuous or in intervals. It is also possible the container comprising the resin beads is simply rocked gently a few times every now and then.

In one embodiment of the invention more than one cell line or type of primary cell is attached to or cultivated on the beads. Hence for example 2, such as 3, for example 4, such as 5, for example 6, such as 7, for example 8, such as 9, for example 10, such as in the range of 10 to 20, for example in the range of 20 to 50, such as more than 50 different cell lines may be attached to or cultivated on said beads. Also different specific primary cells may be attached to the cultivated beads.

It is possible that a subgroup of resin beads only comprise one cell line or a specific kind of primary cells and another subgroup of resin beads comprises another cell line or another specific kind of primary cell and so forth. However, it is also possible that in principle every resin beads comprises all the different cell lines.

Intermediates between these two extremes may also be envisaged. Preferably, said different cell lines and/or primary cells comprise different reporter systems, hence it is possible that the different cell lines are derived from the same parent cell lined by insertion of different reporter systems. However, the different cell lines may also be unrelated.

Specific, non-limited examples of cells, methods of preparing cells comprising reporter systems and methods useful for cultivating cells on resin beads are described in examples 8, 9, 10, 11 , and 12 in PCT/DK2005/000347 and in examples 7, 7a, 8, 9, 10, 11 , 12, 13, 13a, 14, 14a and 16 in PCT/DK2005/000348 as well as in example 2 herein below.

Release of library compounds or of adhesion compound

In one embodiment of the invention the library of test compounds is linked to the resin beads or solid supports via a cleavable linker. Hence in one embodiment of the invention, a proportion of the library members may be released from the resin beads, preferably by cleaving the cleavable linker. The thus released library mem- bers may then interact with cells in the immediate proximity, i.e. normally with cells attached to the same bead, and it is even possible that the library member may enter the cells and interact with intracellular compounds. Resin beads selected based on a desired cellular response may then be incubated with one or more proteomes and proteins of said proteome(s) may interact with library members retained on the resin beads. Later selection of a single bead allows elucidation of the structure of the specific library member remaining attached to said bead.

Preferably, "releasing a proportion of a library member" means releasing one or more copies of the library member attached to a solid support or resin bead. Pref- erably, said copies of the library member are released by cleaving the cleavable linker. For example, in the range of 5 to 95% of all copies of a library member attached to a resin bead are released, more preferably in the range of 5 to 90%, even more preferably in the range of 5 to 80%, such as in the range of 5 to 70%, for example in the range of 5 to 60%, such as in the range of 5 to 50%, for example in the range of 5 to 40%, such as in the range of 5 to 30%, for example in the range of 5 to 20%, such as in the range of 5 to 10% of all copies of a library member attached to a resin bead are released.

Preferably, at the most 50%, more preferably at the most 40%, even more prefera- bly at the most 30%, yet more preferably at the most 20%, such as at the most 10% of all copies of a library member attached to a resin bead are released. In addition it is preferred that at least 1%, more preferably at least 5% of all copies of-a library member attached to a resin bead are released.

It is also comprised within the invention that the adhesion compound may be attached to the resin bead via a cleavable linker. Cleavage of said cleavable linker may release the adhesion compound as well as cells attached to said adhesion compound. When the cleavable linkers linking the library compound and the adhesion compound, respectively, are differentially cleavable, then selective release of either library compound or adhesion compound may be achieved.

The cleavable linker may be any chemical moiety which may be used to attach any molecule to a solid support either covalently or via complex formation, and thereafter release said molecule by the action of either acid, base, electrophiles, nucleophiles, oxidative agents, reductive agents, metals or light. Preferably, the cleavable linker attaches the library member to the solid support covalently. Depending on the nature of the cleavable linker, a person skilled in the art will be capable of controlling cleavage of the cleavable linker, so only a proportion of the copies of a library member are released. A comprehensive review describing state of the art for "cleavable linkers" is "Linkers and Cleavage Strategies in Solid-Phase Organic Synthesis and Combinatorial Chemistry" , F. Guillier, D. Orain, and M. Bradley, Chem. Rev. 2000, 100, 2091 -2157. Any of the cleavable linkers described therein may be used with the present invention.

Examples of useful acid labile linkers include the most commonly used linkers for acidic detachment from a solid support, the Wang and Rink linkers (see Fig. 4A and B in PCT/2005/000347). Detachment of peptide esters from Wang linkers require up to 95% TFA in DCM whereas detachment of Rink esters (Fig. 4B, X=OH) may be cleaved under milder conditions (AcOH - DCM 1 :9) which does not cleave the nor- mal protection groups on the peptides. The Rink amides (Fig. 4A, X=NH2) require 95% TFA (aq). Partial detachment of the compounds attached to the resin may also be achieved using gaseous acids such as HCL or TFA vapour in a sealed container. The use of gases allows rigorous control of the degree of cleavage obtained with concentration of acid and time of exposure. The skilled person may readily establish a suitable concentration of acid and time of exposure to obtain a desried degree of cleavage.

Examples of useful base-labile linkers includesWang and HMBA linkers, which may be cleaved under alkaline conditions. Saponification with 0.1 M NaOH may be ap- plied but even milder conditions such as potassium carbonate in MeOH are applicable. The HMBA linker is stable to TFA under normal conditions.

In a preferred embodiment the cleavable linker is a light sensitive cleavable linker which, upon the action of light with a given wave length and intensity, may release any active compound from the solid support.

Photo-labile linkers provide a tool for solid phase synthesis which enables the detachment of the synthesized molecules in the prescence of acid or base-sensitive functionalities within the molecules. In 1973 Rich proposed the use of o-nitrobenzyl type of linkers (nitrated analogs of the Wang linker). Irradiation with UV-light gave detachment of the free acids or amides although only in moderate yields. Detachment yields could be improved by applying the NBA type linkers (see Fig. 4E of PCT/DK2005/000347). Even better results have been obtained with the Holmes- type linkers (Fig. 4F). Detachment from photolabile linkers is performed by illuminat- ing the resins with ultraviolet light, preferable at 365 nm. The wave length and intensity of the light and the time of exposure might need optimization for the individual case. A person skilled in the art can readily establish conditions wherein a desired proportion of copies of a library member are released. Detachment yields may be over 90 % under ideal conditions.

Example of a preferred photo sensitive cleavable linker:

X = NH₁ 0 ; R = H, alkyl ; R' = linked library member

Depending on the nature of the cleavable linker, the library member may be released using different methods. For example, if the linker is photo labile, the library member may be released by illumination. The release should preferably be partial, so that only a proportion of the library member is released. The person skilled in the art will readily be able to establish the conditions required for partial release using a specific cleavable linker. An example of how to acchieve partial release is given in example 6 of PCT/DK2005/000347. A more preferred example of how to achieve partial release is given in example 3 herein below.

It is also comprised within the present invention that a library member may be linked to a resin bead via different cleavable linker. For example some copies of a library member may be linked to a resin bead via a first kind of cleavable linker, whereas other copies of the same library member may be linked to the resin bead via a second kind of cleavable linker. Preferably, the first kind of cleavable linker is cleavable by another method than the second kind of cleavable linker. By way of example, the first cleavable linker could be acid or base labile, whereas the second kind of cleav- able linker could be photolabile. Thus, some copies of the library member could be released in order to screen for a cellular response, whereas other copies could be retained on the resin beads and released during identification. Thus, releasing a proportion of a library member could be controlled by using different cleavable linkers.

Cell surface molecules

In one embodiment of the invention the methods of the invention involve identification of compounds modulating a cellular response, which is mediated through a cell surface molecule. Hence, the invention, for example may be useful for identifying compounds modulating the activity of a cell surface molecule, preferably a cell surface molecule capable of activating/repressing a signal transduction pathway. Within the context of the present invention the term "signal transduction pathway" should be understood in its common cell biological meaning, i.e. modulation of an intracellular event triggered by a cell surface receptor.

Signal transduction pathways may for example involve steps of phosphorylation, cleavage of proteins, synthesis of cAMP, activation of transcription, inhibition of transcription, change i intracellular Ca²⁺ concentration, change in membrane potential, subcellular relocalisation of cellular components, complex formation of cellular components, degradation of cellular components and/or change in energy metabolism The signal transduction pathway may also be a pathway resulting in modulation of transcription, for example modulation of transcription regulated by a response element, for example a response element selected from the group consisting of of CRE, SRE, TRE and AP-1. In one embodiment of the invention the signal transduction pathway is a pathway resulting in apoptosis.

The cell surface molecule is preferably a protein, more preferably a protein that is accessible from the extracellular surface. Yet more preferably, the cell surface molecule is a cell surface protein receptor (herein also merely designated "receptor"). A "receptor" within the meaning of the present invention, is a molecule, which at least sometimes is localised at the cell surface and which is capable or associating with at least one ligand. The ligand binding site is accessible from the extracellular surface. Frequently, association with said ligand may alter the activity of the receptor.

In a preferred embodiment the cell surface molecule is a G-protein coupled receptor (GPCR). GPCR is a family of receptors coupled to a trimeric G-protein. GPCR to be used with the invention preferably have 7 transmembrane domains. Examples of useful GPCR are given in table 3 in PCT/DK2005/000348.

GPCR may be divided into subfamilies, accordingly the GPCR may selected from the group consisting of GPCR belonging the rhodopsin like family, the secretin family or the metabotropic family, preferably from the group consisting of GPCR belonging the rhodopsin like family or the secretin family.

In another preferred embodiment of the invention the GPCR is coupled to a G- protein, such as G_s, that stimulates adenylate cyclase. In yet another preferred embodiment of the invention the GPCR is coupled to a G-protein, such as Gi, that inhibits adenylate cyclase. In an even further embodiment of the invention the GPCR is coupled to a G-protein, such as GQ, that activates phospholipase C. Examples of GPCRs coupled to G_s, Gi or Go are given in table 3 in PCT/DK2005/000348.

Other receptors than GPCR may be used with the present invention, for example the cell surface molecule may be a receptor selected from the group consisting of receptors belonging to the family of protein kinase coupled receptors and receptors belonging to the family of receptor kinases.

The family of Protein kinase coupled receptors for example includes receptors for cytokines, interferons and HGF. These receptors do not have intrinsic kinase actvity, but are associated with a kinase.

Activation of preferred protein kinase coupled receptors results in activation of AP-1 , i.e. in increased transcription from genes comprising one or more AP-1 sites in their regulatory sequences. This is in particular true for receptors activated by a cytokine. Receptor kinases are receptors having an intrinsic kinase activity. Frequently said activity may be modulated by association of a ligand. The family for example includes receptors for Insulin, NGF, PDGF, FGF, EGF and GH.

Activation of preferred receptor kinases results in activation of SRE, i.e. in an increase in transcription from genes comprising one or more SRE in their regulatory sequences. This is in particular true for receptor kinases activated by growth hormones.

The receptor may also be an orphan receptor, i.e. a receptor for which no ligand has yet been identified. The methods of the present invention may also be useful for identifying ligands of orphan receptors.

The cell surface molecule may in one embodiment of the invention be a channel which is accessible from the extracellular surface, such as a transmembrane channel. Examples of channels are ion-channels, such as Ca²⁺ channels.

Cellular molecules

In one embodiment of the invention the methods of the invention involve identification of compounds modulating a cellular response, which is mediated through an interaction with or between cellular molecules, more preferably through an interaction with or between intracellular molecules. Cellular molecules may be any cellular molecule, such as proteins, polypeptides, DNA, RNA, molecules of non-protein nature or metal-ions. In preferred embodiment, the cellular molecule is a protein or polypeptide. Intracellular molecules are molecules that are not accessible from the extracellular surface of intact cells. Intracellular molecules may for example be proteins, polypeptides, DNA, RNA, molecules of non-protein nature or metal-ions. In one preferred embodiment, intracellular molecules are proteins or polypeptides not accessible from the extracellular surface of intact cells.

In embodiments of the invention wherein the cellular response is mediated through an interaction between intracellular molecules it is preferred that a proportion of the test compounds are released from the resin beads before or simultaneously with screening for resin beads comprising cells meeting at least one selected criterion.

In one embodiment, the cellular response is mediated through an interaction be- tween cellular molecules of a cellular signal transduction pathway. Hence, the invention, for example may be useful for identifying compounds modulating the activity of a signal transduction pathway. Such compounds could for example activate or repress a signal transduction pathways by

^* modulating the interaction between different or the same cellular molecules, ^* modulating the catalytic activity of enzymes,

^* modulating the synthesis or degradation of cellular molecules,

* modulating transcriptional activity,

* modulating the localization or movement of cellular molecules.

^* modulating the level of cellular molecules (i.e. in a specific cellular compartment or on average throughout the whole cell)

Preferably the cellular molecules are proteins or parts thereof or derivatives thereof, more preferably the cellular molecules are proteins. Even more preferably the cellular molecules belong to the classes of: serine/threonine protein kinases; tyrosine protein kinases, protein phosphatases, phospholipid dependent serine/threonine protein kinases, calmodulin dependent serine/threonine protein kinases, mitogenac- tivated serine/threonine protein kinases, cycline dependent serine/threonine protein kinases, transcription factors, structural proteins, protein scaffolds, proteases, such as caspases, metallo-matrix-proteases, rennin, cathepsins, viral proteases, secreta- ses or ADAM family proteases, or hydrolases, nucleases, synthases, isomerases, polymerises, oxido-reductases, ATPases or GTPases.

The cellular molecules are in one embodiment proteins that are known to participate in protein-protein interactions or complex formations. Such proteins can be selected from proteins listed in databases like BIND (Biomolecular Interaction Network Database) http:// bind.ca.

In another embodiment the cellular molecules are proteins, which not necessarily are involved in protein-protein interactions. In one embodiment of the invention the cellular molecules are involved in regulation of apoptosis. Thus the cellular molecules may be proteins or functional fragments thereof involved in regulation of apoptosis. Proteins involved in apoptosis includes for example caspases, inhibitors of apoptosis (IAPs) or Smac. Inhibitors of apoptosis may for example be XIAP, clAP1/BIRC2, ML-IAP/BIRC7, DIAP1 , DIAP2, OPIAP3, clAP2, NAIP, Apollon or Survivin (see also Vaux and Silke, 2005, Nature Reviews, 6:287-297). Thus, in one example proteins involved in protein-protein interaction may be Smac and XIAP or ML-IAP or a Smac binding fragment thereof. Smac binding fragments of XIAP preferably comprises the BIR3 domain of XIAP, whereas Smac binding fragments of ML-IAP preferably comprises the BIR domain. The do- maine structure of IAPs is well described, see for example Wu, G., Chai, J., Suber, T. L., Wu, J.-W., Du, C, Wang, X., and Shi, Y. (2000) Nature 408, 1008-1012; Matthew C. Franklin et al., Biochemistry 2003, 42, 8223-8231 ; and Liston et al. Oncogene. 2003 Nov 24;22(53):8568-80.

Cellular response

The invention relates to methods of identifying compounds modulating, such as acti- vating or inhibiting, a cellular response linked to a reporter system. The reporter system may be any of the reporter systems described herein below. The methods disclosed by the present invention may be used to identify compounds modifying any cellular response, which is or may be linked to a reporter system generating a detectable output. The person skilled in the art will appreciate that the specific methods disclosed herein may be adapted to any such cellular response. Below, non-limiting examples of cellular responses are described.

In one embodiment of the invention, the cellular response is mediated through a cell surface molecule, for example the cellular response may be activation of a receptor. Hence, the cellular response may for example be modulation of a signal transduction pathway within a cell, such as modulation of a signal transduction pathway mediated by a cell surface molecule. By "activation of a receptor" is meant that the receptor is influenced in a manner that it activates downstream signalling events. Accordingly, the methods according to the present invention may be employed to iden- tify agonists or antagonist of a receptor. In another embodiment of the invention, the cellular response is mediated through interaction between cellular molecules, such as intracellular molecules. The cellular molecules may for example be components of asignal transduction pathway, and thus the cellular response may be activation or repression of a signal transduction pathway.

Examples of cellular responses modulated by signal transduction pathways includes:

* Upregulation or downregulation of the level of a member of the pathway

^* Relocalisation of a member of the pathway ^* Complex formation between members of the pathway or between members of the pathway with other cellular compounds

^* Enhanced or reduced transcription from genes regulated by the pathway

^* Modification by for example phosphorylation of a member of the pathway

^* Activation or inhibition of an enzyme of the pathway ^* Degradation of cellular compound(s) due to upregulation or downregulation of the pathway

* Altered secretion of a compound

^* Change in ion-flux

* Morphological changes * Change in viability

The signal transduction pathway may be a pathway modulated by any of the receptors described in the section herein above. Hence, the cellular response may for example be any of the following: * Activation of adenylate cyclase; i.e. increase in adenylate cyclase activity

^* Increased levels of cAMP

^* Increased transcription of genes regulated by a CRE

* Inhibition of adenylate cyclase; i.e. decrease in adenylate cyclase activity

^* Decreased levels of cAMP * Decreased transcription of genes regulated by a CRE

* Increased activity of phospholipase C

^* Increased level of inositol 1 ,4,5-trisphosphate

^* Increased activity of Protein kinase C (PKC)

* Phosphorylation of proteins, which are phosphorylated by protein kinase C Modulation of a signal transduction pathway can for example be monitored by measuring:

^* the enzymatic activity of an enzyme being part of said signal transduction pathway

^* the level of cyclic nucleotides, i.e. cAMP or cGMP ^* the activity of transcription factors

* the level of specific proteins as quantified through standard proteomics techniques

^* the level of inositol or lipid phosphates

* the level of phosphorylation of specific proteins as quantified through standard proteomics techniques ^* the binding between two or more proteins or polypeptides

* the cellular localization of proteins or polypeptides

The enzymatic activity could for example be the enzymatic activity of serine/threonine protein kinases or of tyrosine protein kinases or of protein phosphata- ses or of phospholipid dependent serine/threonine protein kinases or of calmodulin dependent serine/threonine protein kinases or of mitogenactivated serine/threonine protein kinases or of cycline dependent serine/threonine protein kinases or of proteases or of hydrolases or of nucleases or of synthases or of isomerases or of polymerises or of oxido-reductases or of ATPases or of GTPases. The enzymatic activity could also be protease activity, such as caspase activity.

The cellular response may in one embodiment be modulation of transcriptional activity, such as activation or reduction of transcription of one or more genes. In particular, activation or reduction of transcription of genes regulated by a response ele- ment. Said response element could for example be selected from the group consisting of CRE, SRE, TRE and AP-1.

Hence, the cellular response may also be an increased or decreased level of a particular mRNA within a cell.

By the term "regulated by a response element" is meant that transcription is modulated by said response element, however other elements may also modulate transcription of said gene. By the term "activation of response element" is meant increased transcription of genes regulated by said response element and/or operably associated therewith. In another embodiment of the invention the cellular response is:

* change in the intracellular level of a compound; or

* change in the level of a compound within a specific cellular compartment, for ex- ample within the cytoplasm, in the golgi, in the endoplasmatic reticulum, in Iy- sosomes, in endosomes or in the nucleus

The compound may be any compound, preferably a naturally occurring compound. Frequently, the compound is a compound endogenous to the cell. The compound may thus for example be a salt, an ion, a nucleotide or a derivative thereof, a peptide, a saccharide, a lipid or a biomacromolecule. Biomacromolecules includes for example RNA such as mRNA, polypeptides and proteins. An example of an ion is Ca²⁺ and an example of a nucleotide derivative is cAMP.

In yet another embodiment of the invention the cellular response is relocalisation of a compound. Relocalisation may for example be

* concentration of a compound otherwise dispersed in one or more specific locations

* relocalisation from one cellular compartment to another, for example relocalisation from the cellular membrane to the cytoplasma. * relocalisation from one location within a compartment to another location within the same compartment

^* internalisation of an extracellular compound

The compound may be any compound, such as any of the compounds mentioned in the section above. In one preferred embodiment the compound, which is relocalised is a biomacromolecule, such as RNA, polypeptides or proteins. For example, the compound may be a cell surface receptor (receptor). The cellular response may thus be internalisation of said receptor or relocalisation of said receptor from the cellular membrane to the cytoplasma.

In one embodiment of the invention the cellular response is change in the activity of a compound, such as an increase or a decrease in the activity of a compound, such as a cellular protein. Said compound may for example be an enzyme. The cellular response may for example be induction of the activity of a caspase. Preferred cas- pases are Caspase 3 or 7. In another embodiment of the invention the cellular response is change in phosphorylation of a compound.

In another embodiment of the invention the cellular response is formation or disruption of a complex between compounds. Thus, the cellular response may be a change in interaction between two or more cellular molecules, preferably between two intracellular molecules, such as establishment of an interaction between two or more cellular molecules or disruption of an interaction between two or more cellular molecules. The cellular molecules may be any of the cellular molecules mentioned above, however, preferably the cellular molecules are proteins or fragments thereof.

In one embodiment of the present invention the cellular response is induction or facilitation of apoptosis in living cells, such as induction or facilitation of apoptosis in tumour cells, preferably induction of apoptosis. The cellular response may also be induction or facilitation of apoptosis in cells that have undergone an apoptosis promoting treatment. The cellular response may also be induction or facilitation of apoptosis in cells that have undergone an apoptosis inhibiting treatment, In another embodiment of the invention the cellular response is inhibition or reduction of apop- tosis, for example reduction of apoptosis in cells prone to undergo apoptosis or reduction in apoptosis in cells that have undergone an apoptosis promoting treatment.

In another embodiment of the invention the cellular response is change in the concentration of a compound.

The cellular response may also be altered secretion of a compound, such as increased or decreased secretion of a compound. Said compound could for example be a biomacromolecule, such as a protein, a polypeptide, a peptide, a hormone, a cytokine, or the like.

In another embodiment of the invention the cellular response is change in pH in an intracellular compartment, for example in the cytoplasm. In yet another embodiment the the cellular response is a change in a membrane potential, for example a change in membrane potential over the cell membrane or over the mitochondria membrane.

In an additional embodiment of the invention the cellular response is a change in orientation of the cytoskeleton.

In an even further embodiment of the invention the cellular response is change in morphology, such as change in size or shape. The cellular response may also be change in viability (e.g. apoptosis or necrosis) under specific conditions.

The methods according to the invention may also include identification of compounds modulating more than one cellular response, such as 2, for example 3, such as 4, for example 5, such as more than 5 different cellular responses. Said cellular responses may be any of the responses discussed above.

Reporter system

The reporter system to be used with the present invention should be selected according to the particular cellular response. The reporter system should be capable of generating a detectable output.

In some embodiments of the invention the reporter system may be identical to the cellular response. This is in particular true when the cellular response may be detected without the aid of an additional reporter system, for example when the cellular response is an increase/decrease in the level of a compound, relocalisation of a compound, change in membrane potential, change in pH, change in morphology, complex formation between endogenous compounds or the like.

Hence, the reporter system may be a system endogenous to said cells. For example, the reporter system may comprise the endogenous system regulating the intracellular level of an endogenous compound. By way of example, the reporter system may be the endogenous system of a cell regulating the intracellular Ca²⁺ level. For example, the cellular response could be modulation of a signal transduction pathway involving activation of phospholipase C. Phospholipase C may for example be activated by GPCRs coupled to G₀ (see herein above). Activation of phospholipase C in general leads to increase in the intracellular level of Ca²⁺.

In another example, the reporter system comprises the intracellular localisation of an endogenous compound.

However, the reporter system may also be heterologous to the cell, i.e. a reporter system which has been inserted into the cell for example by recombinant tech- niques.

In embodiments of the invention, wherein the cellular response is modulation of transcription from gene(s) regulated by a response element, it is preferred that the report system comprises a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to a response element, the activity of which is modulated by the cellular response.

In embodiments of the invention, wherein the cellular response is modulation of a signal tranduction pathway, the reporter system may comprise a nucleic acid com- prising a nucleotide sequence encoding a detectable polypeptide operably linked to a response element, the activity of which is modulated by said signal transduction pathway.

For example, if the cellular response is modulation of a signal transduction pathway influencing the activity of CRE and/or SRE, then the reporter system may comprise a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to a response element selected from the group consisting of cAMP response element (CRE) and serum response element (SRE). Examples of such signal transduction pathways include the signal transduction pathways modulated by GPCR of the rhodopsin family or secretin family and by protein kinase receptors and receptors belonging to the family of receptor kinases.

By way of example: 1 ) If the cellular response is activation of a signal transduction pathway activated by a GPCR coupled to a G_s (see herein above) that stimulates adenylate cyclase, then the reporter system may be a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to CRE. Activation of said GPCR may then be detected by detection of increased levels of said detectable polypeptide. 2) If the cellular response is activation of signal transduction pathway activated by a GPCR coupled to a Gi (see herein above) that inhib- its adenylate cyclase, then the reporter system may be a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to CRE. Activation of said GPCR may then be detected by detection of decreased levels of said detectable polypeptide.

Similarly, if the cellular response is modulation of a signal transduction pathway that influences the activity of TRE, then the reporter system may comprise a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to TPA response element (TRE). Examples are GPCRs that are linked to activation of Protein Kinase C such as Gq coupled receptors (see herein above).

Similarly, if the cellular response is modulation of a signal transduction pathway that influences the activity of SRE, then the reporter system may comprise a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to SRE. Examples of such signal transduction pathways include the signal tranduction pathways modulated by growth hormones or cytokines through protein kinase receptors and receptors belonging to the family of receptor kinases.

Similarly, if the cellular response is modulation of a signal transduction pathway that influences the activity of AP-1 , then the reporter system may comprise a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to AP-1. Examples of such signal transduction pathways include the signal tranduction pathways modulated by cytokines or growth factors cytokines through protein kinase receptors and receptors belonging to the family of receptor kinases

The detectable polypeptide may be any detectable polypeptide, however preferably the detectable polypeptide is selected from the group consisting of fluorescent proteins and enzymes.

Fluorescent proteins may for example be green fluorescent protein (GFP) and fluo- rescent mutants thereof, such as yellow fluorescent protein (YFP) or cyan fluores- cent protein (CFP). The fluorescent protein can also be a protein complex, e.g. a di- or tetramer of a fluorescent protein, such as dsRed. Enzymes may for example be selected from the group consisting of luciferase, CAT, galactosidase, alkaline phosphatase and beta-lactamase.

In one embodiment of the invention the reporter system may comprise a biolumi- nescent moiety. For example, if the cellular response is relocalisation of a compound, then the reporter system may for example be said compound linked to a Iu- miniscent moiety, such as a fluorescent moeity. Hence, for example if the cellular response is relocalisation of a polypeptide the reporter system may be a chimeric protein made up of said polypeptide and a fluorescent protein, such as GFP, YFP or CFP. In one preferred embodiment said polypeptide may be receptor.

In one embodiment of the invention the reporter system may detect the level of a cellular molecule, such as a protein. This may for example be achieved by quantifying the amount of a compound i.e. an antibody that specifically binds to the cellular molecule. The quantification can for example be achieved by covalently coupling a fluorescent, bioluminescent or coloured moiety to said compound. The quantification could be confined to a specific cellular compartment.

In one embodiment of the invention the reporter system may detect the level of modification of a cellular molecule for example but not limited to phosphorylation, glycosylation or ubiquitination. This may for example be achieved by quantifying the amount of a compound i.e. an antibody that specifically binds to the modified cellular molecule. The quantification can for example be achieved by covalently coupling a fluorescent, bioluminescent or coloured moiety to said compound. The quantification could be confined to a specific cellular compartment.

In one embodiment of the invention the reporter system may detect complex forma- tion or disruption between two cellular proteins (designated first protein and second protein in this paragraph). This is in particular relevant when the cellular response is change in interaction between two cellular proteins. In this embodiment the reporter system may be endogenous, i.e. it may comprise or consist of cellular proteins capable of complex formation. The reporter system may also be heterologous. Several differerent reporter systems may be used to detect interaction between a first and a second protein. Below preferred reporter systems are described, however, the invention is not limited to these specific reporter systems.

The reporter system may for example comprise the first protein linked to a biolumi- nescent moiety, such as luciferase and the other protein linked to a fluorescent moiety, such as a fluorescent protein. Such reporter systems are referred to as "BRET reporter systems" herein. The bioluminiscent moeity should preferably be able to directly or indirectly generate light of a wavelength capable of exciting the fluorescent moiety. The skilled person will readily be able to select useful bioluminiscent moeities and fluorescent moeities. Preferably, the BRET reporter system comprises a first chimeric protein comprising the first protein linked to a bioluminescent protein, preferably luciferase and a second chimeric protein comprising the second protein linked to a fluorescent protein. Such a reporter system may be introduced into a cell by introducing nucleic acids encoding the first and the second chimeric proteins un- der control of suitable promoters into said cell. Direct interaction between the proteins can after expression of the two chimeric proteins be detected through occurrence of BRET (Bioluminescence Resonance Energy Transfer). In one embodiment, BRET2 technology may be used which is based on energy transfer between a bioluminescent donor (a Renilla luciferase (Rluc) fusion protein) and a fluorescent ac- ceptor (a Green Fluorescent Protein (GFP2) fusion protein). In presence of its substrate DeepBlueC™ (a coelenterazine derivative), Rluc emits blue light (~395 nm). Thus the reporter system may comprise a first chimeric protein comprising the first protein and Rluc and a second chimeric protein comprising the second protein and GFP2. A protein-protein interaction between Rluc and GFP2 chimeric proteins al- lows energy transfer to GFP2, which reemits green light (510 nm). Expression of

Rluc alone, in the presence of the substrate DeepBlueC™, gives an emission spectrum with a peak at ~395 nm, whereas when the Rluc and GFP2 chimeric proteins interact , there is efficient energy transfer between Rluc and GFP2 and the 510 nm signal represents a major peak.

In another similar reporter system the first protein and the second protein are linked to different fluorescent moieties, preferably a fluorescent proteins. Such reporter systems are referred to as "FRET reporter systems" herein. Preferably, one fluorescent moiety is capable of emitting light of a wavelength capable of exciting the other fluorescent moeity. FRET reporter systems preferably comprise a first chimeric pro- tein comprising the first protein and a fluorescent protein and a second chimeric protein comprising the second protein and another flourescent protein. It is then possible to detect the complex formation through the occurrence of FRET (Fluorescence Resonance Energy Transfer). BRET or FRET according to the present inven- tion may for example be performed as described in (Nicolas B, R Jockers, and T lssad Trends in Pharmacological Sciences 23 (8):351-354, 2002; and/or A. Roda, M. Guardigli, P. Pasini, and M. Mirasoli. Anal.Bioanal.Chem 377 (5):826-833, 2003)

Complex formation may also be detected by proximity ligation. In such an embodi- ment the reporter system comprises two cellular compounds. These compounds may be detected using two affinity probes raised against the first and the second protein. Such reporter systems are designated "proximity ligation reporter systems" herein. When the two proteins come in close proximity a ligation reaction creates a DNA reporter sequence that can be amplified. The amplified sequence can be de- tected by any useful method, for example it may be detected through photolabelling. Preferably, the DNA reporter sequence is amplified by PCR, rolling circle replication or ligation chain reaction. In order to detect the amplified sequence, the sequence may for example be amplified using primers labelled with a detecable label, such as a fluorescent label or the sequence may be detected using a detectably labelled probe, such as a f luorescently labelled probe. The affinity probes in general comprise or consist of a binding moeity and a nucleic acid moeity. The binding moiety of the affinity probes can be any molecule that binds either the first or the second protein with high affinity. Preferably, the binding moeity is capable of specifically recognising and binding either the first or the second protein. Examples of usefuil binding moieties of affinity probes are monoclonal- or polyclonal antibodies or antigen binding fragments thereof, chimeric antibodies, recombinant antibodies, single chain antibodies or aptamers. Antibodies may be prepared using any conventional method known to the person skilled in the art. Aptamers may be prepared by any method known to the skilled person, for example by iterative cycles of screening nucleic acid libraries for compounds capable of binding a tare nucleic acid molecules selected for their ability to specifically bind a target. Aptamers may for example be produced using a SELEX process (Sun S Curr Opin MoI Ther 2000 Feb 2:100-5; Jayasena SD Clin Chem 1999 Sep 45:1628-50). The nucleic acid moeity may comprise or consist of any nucleic acid sequence, preferably a sequence, which when ligated to another nucleic acid moeity creates a DNA repoter sequence, which can be amplified by PCR, rolling cycle amplification or ligase chain reaction using appropriate primers. A person skilled in the art can design useful nucleic acid moeity sequences and corresponding primers. The affinity probes can be introduced into cells by a number of different methods, for example they may be introduced into the cells after said cells have been fixed and permeabilized or they can be introduced by using traditional cDNA transfection methods, for example by using standard procedure for Fugeneθ transfection. Proximity ligation may for example be carried out as described in Frederiksson et al. Nature Biotechnology 2002, 20: 473; Gullberg et al. Curr Opinion Biotechnology 2003, 14: 82. The reporter system may also be a "two-hybrid reporter system". Two-hybrid reporter systems comprises two chimeric proteins, wherein the first chimeric protein comprises the first protein fused to a DNA binding domain and the second chimeric protein comprises the second protein fused to a transactivating domain. Furthermore, the two hybrid reporter system comprises a reporter construct comprising a nucleic acid sequence encoding a detectable polypeptide the expres- sion of which is controlled by the transactivating/DNA binding domain. Thus if the first protein interacts with the second protein the DNA binding domain and the trans- activating domain are brought into close proximity and may activate transcription from the reporter construct. Interaction can then be determined by detection of the detectable polypeptide. The detectable polypeptide may be any of the detectable polypeptides mentioned herein above. Two-hybrid reporter systems are well described in the art, see for example US 5,283,173.

The reporter system may also be an enzyme complementation reporter system. Enzyme complementation reporter systems comprises two chimeric proteins, wherein the first chimeric protein comprises the first protein fused to a first part of an enzyme and the second chimeric protein comprises the second protein fused to a second part of an enzyme. The first and the second part of an enzyme should together constitute a functional enzyme. Thus, when the first protein interacts with the second protein the first part and the second part of an enzyme will form a functional enzyme, the activity of which may be determined. One example of an enzyme useful for enzyme complementation system is DHFR (dihydrofolate reductase), where the activity of the reconstituted enzyme is monitored as a fluorescent read-out based on stoichiometric binding of fluorescein-methotrexate to reconstituted DHFR (Remy I, Michnick SW Proc Natl Acad Sci U S A 1999 May 96:5394-9). When the cellular response is induction/facilitation of apoptosis a number of reporter systems may be employed. Induction/facilitation of apoptosis may be determined by determing caspase activity as described herein above. Induction/facilitation of apoptosis may also be determined by determining cell growth/number of cells, for exam- pie cell growth/number of cells after cultivation of for example normal cells or tumour cells, such as cells expressing high levels of an inhibitor of apoptosis (IAP) such XIAP or ML-IAP. Induction/facilitation of apoptosis may also be determined byu determining membrane integrity, for example as described in example 3. Other methods of determining apoptosis are well known to the skilled person.

Specific, non-limited examples of useful reporter systems and methods for screening and selecting resin beads based on these reporter systems are described in examples 8, 9, 10, 11 , and 12 in PCT/DK2005/000347 and in examples 7, 7a, 8, 9, 10, 11 , 12, 13, 13a, 14, 14a and 16 in PCT/DK2005/000348. In addition, example 3 herein below describes a non-limiting example of a useful reporter system.

Detectable output

The detectable output may be any output, which is detectable directly or indirectly. For example the detectable output may be the concentration of a compound within a cell, localisation of a compound within a cell, luminiscense, activity of an enzyme or the like.

In preferred embodiments of the invention the detectable output is luminiscense, such as fluorescence, bioluminescence, FRET or BRET. Bioluminiscence may be detected by any conventional methods, for example with the aid of a Plate reader. BRET may be performed as described herein above. In one embodiment, BRET2 technology is used which is based on energy transfer between a bioluminescent donor (a Renilla luciferase (Rluc) fusion protein) and a fluorescent acceptor (a Green Fluorescent Protein (GFP2) fusion protein). In presence of its substrate DeepBlueC™ (a coelenterazine derivative), Rluc emits blue light (-395 nm). A protein-protein interaction between Rluc and GFP2 fusion proteins allows energy transfer to GFP2 which reemits green light (510 nm). Expression of Rluc alone, in the presence of the substrate DeepBlueC™, gives an emission spectrum with a peak at ~395 nm (solid line). With the Rluc:GFP2 fusion construct, there is efficient energy transfer between Rluc and GFP2 and the 510 nm signal represents a major peak (dashed line). The BRET2 signal is expressed as the 515 nm to 410 nm ratio, since filters centered at those wavelengths are used for detection. FRET technology is based on the distance-dependent energy transfer between two fluorescence groups that are each coupled to a protein.

Alternatively, the detectable output may preferably be linked (directly or indirectly) to a bioluminiscent signal.

However, the detectable output could also be radioactivity, a coloured compound or a colour signal, a heavy metal, an electrical potential, a redox potential, a temperature or the detectable output may be linked to a radioactive signal, a coloured compound or a colour signal or a heavy metal or an electrical potential, or a redox poten- tial or a temperature. Said radioactive signal could for example be 35S, 32P, 3H. The coloured compound could for example be the product of any of the enzymatic reaction described herein elsewhere. The heavy metal could for example be gold.

In embodiments of the invention, wherein the cellular response is change in the in- tracellular level of a compound or change in the level of a compound within a specific cellular compartment, then the detectable output may be said level of said compound. Depending on the nature of the compound, said level may be detected directly or indirectly.

If the compound for example is a fluorescent compound, the level of said compound may be determined by determining the fluorescence properties. This may be done by any suitable means, for example by the aid of a fluorescence microscope, a FACS (Fluorescence Activated Cell Sorter), a FABS (Fluorescence Activated Bead Sorter), fluorescence plate-reader or a fluorescence spectrometer,

If the compound for example is an enzyme then the level of said compound may be determined by determining the activity of said enzyme. By way of example, if the enzyme catalyses a reaction leading to a product, which is directly detectable, for example by colorimetric or chemiluminescent detection techniques, the activity of said enzyme may be detected by detecting said compound. For example, if the en- zyme is luciferase, the activity of said enzyme may be detected by detecting em- mision of light upon oxidation of the added substrate, luciferin.

Several other enzymes such as CAT, (-galactosidase, alkaline phosphatase, horse- radish peroxidase and beta-lactamase are, when provided with suitable substrates, capable of catalysing reactions leading to coloured or chemiluminescent products, which may be detected using any colorimetric or chemiluminescent detection technique.

If the compound for example is Ca²⁺, then the intracellular concentration of said ion can be measured by using any suitable method, for example by inserting into the cells Ga²⁺ binding fluorescent compounds like Fura-2, Fluo-3 or Fluo-4 (Molecular Probes), which change fluorescent properties according to a changed Ca²⁺ concentration. Non-limiting examples of methods of determining cytosolic free Ca²⁺ are given in examples 13 and 13a of PCT/DK2005/000348. Other ion concentrations can be monitored using suitable fluorescent compounds, which for example are available from Molecular Probes Inc.

If the compound for example is a protein, then it may for example be detected using a first specific binding partner. Said first specific binding partner could be a second protein capable of specifically interacting with said protein, such as a specific antibody or said first specific binding partner could be an aptamer. Said first specific binding partner could be conjugated to a directly detectable compound, such as a fluorescent compound, a radioactive compound or a heavy metal or to an indirectly detectable compound, such as an enzyme, which for example could be any of the enzymes mentioned herein above. It is also possible that the first specific binding partner may be detected with a second specific binding partner, capable of interacting specifically with the first specific binding partner. Said second specific binding partner may be conjugated to a directly or indirectly detectable compound similarly to the first specific binding partner. Additional specific binding partners may be used.

In embodiments of the invention wherein the cellular response is relocalisation of a compound the detectable output could be a detectable label conjugated to said compound. In particular, the compound may be conjugated to a directly detectable label, such as a fluorescent label or a heavy metal. Thus the localisation of the com- pound may be directly detected, for example using a fluorescence microscope, Fluorescent plate-reader, fluorescence spectrometer, a FACS or a FABS instrument In one preferred embodiment the compound is a fusion protein comprising a protein of interest and a fluorescent protein, such as GFP. The compound may thus be a fluo- rescent probe. Thus the detectable output may be localisation of a fluorescent signal. Alternatively, the compound is a fusion protein comprising the protein of interest and a tag. Said tag could be a tag specifically interacting with a specific binding partner, for example the tag could be an HA-tag or a flash domain. Alternatively, localisation of a compound may be determined with the aid of a specific binding partner as outlined above. Intracellular localisation may also be detected using methods capable of detecting distance between two compounds, for example BRET or FRET.

In embodiments of the invention wherein the cellular response is change of activity of a compound, the detectable output may be a product of said activity. I.e. when said compound is an enzyme the detectable output could be a product of a reaction catalysed by said enzyme. Said product could thus be a coloured product or a chemiluminiscent product as discussed herein above.

In embodiments of the invention wherein the cellular response is enhanced or reduced transcription from one or more genes, then the cellular response could be mRNA transcribed from said gene, a protein encoded by said gene or in case the protein is an enzyme, the detectable output could be a product of a reaction catalysed by said enzyme. The enzyme and the products could be any of the enzymes or products discussed herein above.

mRNA may be detected by any useful means, for example with the aid of a probe capable of hybridising specifically with said mRNA. Said probe could be labelled with a directly detectable label, for example a radioactive compound, a fluorescent compound or a heavy metal or an indirectly detectable label such as an enzyme or a specific binding partner.

Said protein may be detected with the aid of specific binding partners as outlined herein above. However, in a preferred embodiment the protein is a fluorescent pro- tein and may thus be detected directly. Hence, the detectable output could be bio- luminescence, such as fluorescence.

In embodiments of the invention wherein the cellular response is modification by for example phosphorylation of a compound this can be detected through binding of a antibody that specifically bind the phosphorylated protein said antibody can then be quantified by specific fluorescence labelling.

In embodiments of the invention wherein the cellular response is change in pH in an intracellular compartment, the detectable output will in general be said pH. The pH may be determined using any suitable method, for example using a pH indicator or a pH-meter. For example the pH may be determined using a fluorescent indicator for intracellular pH. Suitable compounds are compounds with a fluorescent excitation profile which is pH-dependent, such as BCECF (available from Molecular Probes). In embodiments of the invention wherein the cellular response is a change in a membrane potential, the detectable output will in general be said membrane potential. The membrane potential may be determined using any suitable method such as applying a fluorescent molecule to cells that distribute over the membrane dependent upon the membrane potential. Examples of such compounds are DiBAC, various ANEP dyes, JC-1 and JC-9 (Molecular Probes). For example, JC-1 and JC-9 are cationic dyes that exhibit potential-dependent accumulation in mitochondria leading to a shift in fluorescence emmision from green to red. Thus mitochondrial depolarization may for example be determined by decrease in red/green fluorescence intensity ratio (see also product information from Molecular Probes). ANEP dyes are in particularly useful for detection of changes in membrane potential. The fluorescence can be readfor instance by a fluorescence microscope, a fluorescence plate-reader, a FACS, or a FABS instrument.

In embodiment of the invention wherein the cellular response is change in morphol- ogy, the detectable output will in general be the morphology of the cell. The morphology may be observed using any suitable method for example by the aid of a microscope, using a FACS or FABS.

In embodiments of the invention wherein the cellular response is change in cy- toskeletal orientation, the detectable output may for example be a morphological change, such as a morphological change of the fluorescence observed after staining of actin with phalloidin. This change in fluorescence may be observed using any suitable method for example by the aid of a microscope, using a FACS or FABS.

In embodiments of the invention where the cellular response is change in an interaction between two cellular proteins, the detectable output may for example be BRET or FRET, which is detectable by determining the occurrence of fluorescence of a given wavelength. BRET or FRET may for example be detected using a FABS, FACS₁ fluorescent microscope or any other equipment useful for detection of fluo- rescence.

In embodiments of the invention wherein the reporter system is proximity ligation the detectable output is dependent on the detectable label used to label the amplified DNA reporter sequence. In embodiments, wherein the DNA reporter sequence is amplified using fluorescently labelled primers, the detectable output will be said fluorescent label, which may be detected using a FABS, FACS, fluorescent microscope or any other equipment useful for detection of fluorescence.

Depending on the detectable output, it will frequently be an advantage to fix cells prior to detecting said detectable output. However, in some embodiments of the invention it is preferred that the cells are not fixed. Cells may be fixed according to any useful protocol (see also definitions herein above).

Selection related to cellular response

The methods according to the invention involves screening resin beads for beads comprising cells meeting at least one predetermined selection criterion and wherein said selection criterion is linked directly or indirectly to said detectable output. Hence, the selection criterion will be dependent on the detectable output.

For example the predetermined selection criteria may be a quantitative criterium, such as a quantitative level of bioluminiscence above or below a specific threshold value. In embodiments of the invention, wherein the detectable output is fluorescence or the detectable output may be linked to a fluorescent signal, then the predetermined selection criterion could be any fluorescence property. For example, the selection criterion could be intensity of said fluorescence above or below a predetermined threshold value or emission of light of a specific wavelength or absorption of light of a specific wavelength or intensity of emitted light of a specific wavelength above or below a predetermined threshold value. The selection criterion could also be based on Fluorescence lifetime and/or fluorescence polarization The selection criterion could also be a specific localisation of the fluorescent signal, such as intensity of a fluorescent signal in a specific cellular compartment above or below a predetermined threshold value. The selection criterion could also be a predetermined change in fluorescence lifetime or in fluorescence polarization. Fluorescence intensity and/or localisation may for example be determined using image processing and/or image analysis, a fluorescence microscope, FACS, FABS or fluorescence plate reader.

In one embodiment of the invention the selection criterion is high fluorescence intensity. This may for example be the case, when the cellular response is activation of a signal transduction pathway and the reporter system comprises a gene encod- ing a fluorescent protein, where activation of the signal transduction pathway leads to incresed expression of said gene. Then resin beads may be selected using a method comprising the steps of:

1. Determining the fluorescence intensity of positive control resin beads and setting this fluorescence intensity to 100% 2. Determining the fluorescence intensity of negative control resin beads and setting this fluorescence intensity to 0%

3. Selecting resin beads having a fluorescence intensity corresponding to at least 5%, preferably at least 10%, more preferably at least 20%, even more preferably at least 30%, such as at least 40%, for example at least 50%, such as at least 60%, for example at least 70&, such as at least 80%, for example at least 90%, such as in the range of 5 to 100%, for example in the range of 10 to 100%, such as in the range of 20 to 100%, for example in the range of 30 to 100%, such as in the range of 40 to 100%, for example in the range of 50 to 100%. The positive control may for example be a resin bead (or optionally several resin beads kept in a separate container or well) comprising a compound known to influence the cellular response. By way of example, if the cellular response is activation of a signal transduction pathway through a cell surface receptor, then the positive control may be a resin bead comprising a known ligand of said receptor, for example a naturally occurring ligand. The negative control may be a resin bead (or optionally several resin beads kept in a separate container or well) optionally comprising a cell adhesion compound, but otherwise comprising no library member or other test compound.

In another embodiment of the selection criterion is low fluorescence. This may for example be the case, when the cellular response is inhibition of a signal tranduction pathway and the reporter system comprises a gene encoding a fluorescent protein, where an active signal transduction pathway leads to expression of said gene. Then resin beads may be selected using a method comprising the steps of:

1. Determining the fluorescence intensity of positive control resin beads and setting this fluorescence intensity to 0%

2. Determining the fluorescence intensity of negative control resin beads and setting this fluorescence intensity to 100% 3. Selecting resin beads having a fluorescence intensity corresponding to at least 5%, preferably at least 10%, more preferably at least 20%, even more preferably at least 30%, such as at least 40%, for example at least 50%, such as at least 60%, for example at leasat 70&, such as at least 80%, for example at least 90%, such as in the range of 5 to 100%, for example in the range of 10 to 100%, such as in the range of 20 to 100%, for example in the range of 30 to 100%, such as in the range of 40 to 100%, for example in the range of 50 to 100%.

The positive control may for example be a resin bead (or resin beads) comprising a compound known to influence the cellular response. By way of example, if the cellu- lar response is inhibition of a signal transduction pathway through a cell surface receptor, then the positive control may be a resin bead comprising a known antagonist of said receptor. The negative control may be a resin bead (or resin beads) optionally comprising a cell adhesion compound, but otherwise comprising no library member or other test compound. In one preferred embodiment selection is performed manually with the aid of a fluorescence microscope. In this embodiment the fluorescence intensity or other fluorescence properties are judged manually.

When the selection criterion is fluorescence intensity of localisation, the resin beads may also be analysed using a plate reader or image acquisition.

If the selection criterion is localisation, then resin beads are generally analysed by a fluorescence or imaging microscope. Said microscope may optionally be equipped with a micromanipulator capable of picking out single beads. Resin beads are scanned for cells where the fluorescence signal is located at the desired intracellular location and these resin beads are selected. The selection may be manually or it may be automated.

In embodiments of the invention, wherein the detectable output is light emission or the detectable output may be linked to a light signal, then the predetermined selection criterion could be any property of the light. For example the selection criterion could be light intensity above or below a predetermined threshold value. Light can be detected for example by the eye, in a microscope, and if the light is emitted via bioluminescence it can be measured by a luminometer.

In embodiments of the invention, wherein the detectable output is a radioactive signal or the detectable output may be linked to a radioactive signal, then the selection criterion could be any property of said radioactive signal, such as intensity above or below a predetermined threshold value or localisation of the radioactive signal.

In embodiments of the invention, wherein the detectable output is a colour signal or the detectable output may be linked to a colour signal, then the selection criterion could be any property or said colour signal. For example the predetermined selec- tion criterion could be a colour intensity above or below a specific threshold value or it could be a specific colour. The colour signal could be detected using any suitable colorimetric method, such as a spectrophotometer,

Resin beads comprising cells meeting at least one selection criterion, such as any of the selection criteria mentioned herein above are selected. In certain embodiments of the invention resin beads comprising cells meeting at least two, for example 2, such as 3, for example 4, such as in the range of 5 to 10, for example of in the range of 10 to 25 selection criteria are selected.

It is also possible within the present invention to select resin beads comprising cells meeting one or more predetermined selection criteria and subsequently to subject said beads to one or more additional selection rounds, wherein resin beads comprising cells meeting one or more additional selection criteria are selected.

Resin beads meeting said at least one predetermined selection criteria may be selected by manually sorting for example with the aid of a microscope, for example by sorting by fluorescence or by colour or by morphology depending on the detectable output and the selection criterion. Positive beads may be picked directly under the microscope, such as under a fluorescence microscope for example manually or with the aid of a micromanipulator. Frequently, in the range of 100 to 1 ,000,000, for example in the range of 1000 to 100,000, such as in the range of 5000 to 50,000 resin beads may be placed on a suitable surface, such as in a dish or on a coverglass and subsequently examined by microscopy. Alternatively, the sorting process may be automated with the use of specially designed, commercially available bead sort- ers (Union Biometrica, Sommerville, Mass.) and detecting for example fluorescence intensity (Meldal, 2002, Biopolymers, 66: 93-100). In general, resin beads can be sorted at a rate of about 100 beads per second, or even faster depending on the equipment used and its reading capacity. A range of about 5-30 beads per second is generally used with known instruments. Slower rates may be used to increase accuracy, however any suitable rate may be used with the present invention, such as much higher rates. Preferred, is a rate where only one resin bead passes through the detector at a time. It is also comprised within the present invention to select resin beads using a plate reader. In general in the range of 1 to 1000, such as 10 to 500, for example 50 to 100 resin beads are placed in each well of a micro titre plate and analysed. Beads from positive wells may then be further examined.

In one embodiment of the invention resin beads may be selected by comparing the detectable output, with the detectable output generated by control resin beads, for example positive and/or negative control resin beads. Positive control resin beads are beads comprising a compound capable of inducing the desired cellular re- sponse, whereas negative control resin beads comprise no such compound. By way of example, if the cellular response is activation of a cell surface receptor with a known natural ligand, the positive control resin bead may comprise said ligand, whereas the negative control resin bead comprises no compound except optionally a cell adhesion compound.

If the detectable output is a quantifiable signal, then resin beads may be selected, comprising ceils where the detectable output is higher or lower than the detectable output from cells attached to the positive or negative control resin bead. By way of example, if the detectable output is fluorescence intensity, then resin beads comprising cells displaying a fluorescence intensity which is higher than the negative control and lower than the positive control could for example be selected.

After selection of resin beads, cells will in general be released from the resin beads. This may be performed by cleavage of the cleavable linker linking the adhesion compound to the resin beads as described herein above in the section "Release of library compounds or of adhesion compound".

After release of cells the resin beads are ready for incubation with a proteome. In embodiments of the invention wherein the resin beads have been incubated with a proteome prior to screening for a cellular response, then the resin beads are ready for identification of the library member attached to said resin beads.

Multiplexing

The methods disclosed by the present invention may also be used in multiplexing methods.

For example, the methods may be used to identify compounds modifying at least two cellular responses, such as 2, for example 3, such as 4, for example in the range of 5 to 10, such as in the range of 10 to 25 cellular responses.

In such methods step c) of the method outlined above (see the section "Summary of the invention") preferably involves screening resin beads for beads comprising cells meeting at least two, such as 2, for example 3, such as 4, for example in the range of 5 to 10, such as in the range of 10 to 25 predetermined selection criteria, wherein each selection criterion is preferably related to a different detectable output.

In such a method more than one kind of cell may be attached to each resin bead and the different cellular responses may be detected in different kinds of cells. For example, a first cell line comprising a first reporter system linked to a first cellular response and a second cell line comprising a second reporter system linked to a second cellular response and optionally additional cell line(s) comprising additional reporter system(s) linked to additional cellular response(s) may all be attached to a single bead. Resin beads comprising cells meeting selection criteria linked to all the different reporter systems may then be selected.

Depending on the detectable outputs, said detectable output may be determined using any of the methods described herein above. In one preferred embodiment at least two detectable outputs are fluorescent outputs, preferably of different excitation and/or emmision. Thus resin beads meeting said at least two selection criteria may be selected in one step using a FABS with at least 2 channels in both excitation and emmision. Similarly, more than two different fluorescent properties may be selected for in an suitable FABS. The at least two detectable outputs may be in the same cell line or they may be in different cell lines.

Proteome

Protein mixtures to be used with the present invention may be derived from a variety of different sources. Protein mixtures to be used with the present invention should comprise at least 2, preferably at least 100, more preferably at least 200, even more preferably at least 300, such as at least 500, for example at least 1000 different pro- teins. In general, the protein mixture will be derived from one or more natural sources, such as for example from cells, such as living cells, from pathogens, from tissues, from entire individuals, from body fluids such as urine, sputum, serospinal fluid, serum or blood, or from an extracellular matrix. Furthermore, protein mixtures may be obtained from culture media used for cultivating cells, such as bacterial, fungal, plant or animal cells, such as mammalian cells. The protein mixture can be obtained from any source, including, for example, simple organisms, for example pathogens such as fungi, viruses, protozoans and bacteria to more complex organisms such as plants and animals, including mammals and particularly, humans. The biological material may be extracted from individual cell lines, from cellular organisms, or from tissue containing a large variety of cell types or from entire multicellular organisms. Protein mixtures may be prepared for example from healthy cells or tissues or from diseased cells or tissue. Diseased cells or tissue may for example be established cells lines derived from a tumour, tumour cells or even tumour tissue removed for example by surgery.

In a preferred embodiment at least one protein mixture is a mixture of mammalian proteins, preferably human proteins, such as mammalian (human) tissue cell proteins. Protein mixtures may also comprise recombinantly engineered proteins, for example the protein mixtures may also be obtained from cellular systems expressing a cDNA library that may be tagged, for example, with a genetic label that is co- expressed and used for detection analysis. Suitable genetic tags include, for example, myc and photoproteins such as Green Fluorescent Protein (GFP). Alternatively, the protein mixture may comprise proteins encoded by mutagenised, recombined or otherwise manipulated nucleic acids.

In some embodiments of the present invention more than one different protein mixtures is applied, for example 2, such as 3, for example 4, such as 5, for example in the range of 5 to 10, such as more than 10 different protein mixtures. Thus in some embodiments of the invention at least 2 different protein mixtures are used. Preferably, said at least 2 different protein mixtures are mixtures that are desirable to compare. For example, the protein mixture may be mixtures derived from a healthy and a diseased population, respectively, protein mixtures derived from different organisms, protein mixtures derived from different species, protein mixture derived from different tissues, protein mixtures derived from differentially developed organisms or protein mixtures derived from cells or organisms in different states, i.e. cycling versus non-cycling cells. Diseased populations include cells/body fluids/tissues derived from diseased tissue, body fluid or cells derived from an individual with a disease. Said disease could for example be a neoplastic or preneoplastic disease, an infec- tious disease, an autoimmune disease, a cardiovascular disease, an inflammatory disease, CNS disorders, metabolic diseases or endocrine diseases. In one embodiment of the present invention at least one protein mixture is derived from an infectious species, such as for example fungi, viruses, protozoa or bacteria. If for example protein mixtures derived from a healthy and a diseased source, respectively are used, the methods may be used to identify ligands capable of specifically interacting with diseased or healthy cells or tissue may be identified. Such ligands may be potential drug candidates. If for example protein mixtures derived from different species, wherein one species is an infectious agent, ligands interacting specifically with said infectious agent may be identified. Such ligands may also be potential drug candidates.

In another embodiment of the invention the protein mixture comprises one or more families of proteins, which enables detection of ligand-protein binding pairs by immunoassays, for example by the aid of antibodies recognising said families of pro- teins.

The protein mixtures may be obtained from above-mentioned sources by a number of methods well known to the skilled person. Thus, protein mixtures may be extracted and optionally further purified according to conventional protocols. There are various known ways of isolating proteins from cells, tissue, and organisms while preserving the activity of the proteins. The proteins can for example be extracted and solubilized using a variety of auxiliary substances such as detergents and ureas. This extraction procedure is particularly important for larger, hydrophobic proteins such as membrane proteins. The use of detergents, ureas, and salt is com- patible with screening on solid phase resins. Proteins can be extracted using standard equipment such as the French Press and/or a sonicator. The extraction procedure can be manipulated to enrich for low abundance proteins or to isolate a particular class of proteins. General protocols for the extraction of proteins from different organisms are readily available. See, for example, 2-D Proteome Analysis Proto- cols, A.J. Link (Ed), 1st Ed, 1999, Humana Press: Totowa).

Non-limiting specific examples of methods for preparing protein mixtures are described in WO2004/062553 in examples 10, 11 , 12, 13 and 46. Detecting and isolating ligand-protein binding pairs

A variety of suitable methods are useful for detecting the ligand-protein binding pairs. For example, where a single protein mixture is used, the extracted protein may be immediately incubated with the solid supports selected after screening for modulation of a cellular response. Alternatively, in embodiments of the invention the single protein mixture may be incubated with the solid support attached the library of test compounds. After washing, bound protein can be detected directly in the binding complex by the application of a detection molecule to the incubation mixture. The detection molecule is preferably directly detectable and capable of binding proteins. Examples of useful detection molecules include silver or fluorescent dyes that do not interact with the ligand or the solid support.

In another embodiment protein binding is detected by washing the resin beads with a denaturing solution, running said denatured solution on a 1 D or 2D gel, such as by SDS-PAGE and detecting proteins using useful detection molecules such as silver, fluorescent dyes, coomassie and the like. This is in particular useful when a single, non-labeled protein mixture is directly incubated with the resin beads.

It is however generally preferred that the protein mixture is labeled with a detection probe prior to incubation with the solid supports. Hence, in another embodiment, the mixture of proteins may be labeled with a detection probe, for example, with a fluorescent dye such as Oregon Green 514 (green), N-mehtyl anthranilate (blue), Rho- damine red (red), cyanine dye 2, cyanine dye 3, cyanine dye 5 or other commonly used fluorescent probes. See for example, Richard P. Haugland, "Handbook of Fluorescent Probes and Research Products", 9th Edition, 2002, Molecular Probes Europe BV: Leiden or world wide web (WWW) sites "probes.com" and "amersham- biosciences.com/aptrix/upp00919.nsf/Content/DrugScr+CyDye+Fluors+introduction" for a description of cyanine fluorescent dyes. The detection probe may also be a fluorescent protein, such as Green fluorescent proteins or fluorescent mutants thereof. The detection probe can also be a probe that produces chemolumines- cence, such as luciferase or aequorin. In these embodiments, after incubation of ligands with proteins, the library is washed and ligand-protein binding complexes will be detected via the label, for example, fluorescence or color. These ligand-protein binding pairs can be immediately isolated using automatic or manual sorting proce- dures. If the detection probe is a fluorescent probe, then automatic sorting preferably involves the use of a FABS and/or a fluorescence activated beads sorter. Manual sorting may for example involve the use of a fluorescence microscope. The detection probe may furthermore be a compound capable of producing chemilumines- cence, such as for example luciferase or aequorin. The detection probe may furthermore be an enzyme capable of catalyzing a detectable reaction, such as for example phosphatase or peroxidase. The detection probe may furthermore be a metal, for example gold. The protein mixture may be labeled with the detection probe by any conventional method depending on the nature of the detection probe.

In particular, when more than one protein mixture is used it is preferred that the protein mixtures are labeled with a detection probe prior to incubation with the solid supports. The individual protein mixtures may be labeled using different detection probes or similar detection probes. It is however preferred that different protein mix- ture are labeled with different detection probes, to allow identification of from what protein mixture the protein is derived. Hence, it is preferred that 2, such as 3, for example 4, such as 5, for example in the range of 5 to 10 differentially labeled protein mixtures are used with the invention. Selecting an immobilised ligand-protein binding pair may thus involve detectecting a compound of the library that binds dif- ferentially with 2 or more differentially labeled protein mixtures.

It is comprised within the present invention that individual proteins of a protein mixture may be differentially labeled, it is however preferred that all proteins of one protein mixture are labelled with the same kind of detection probe. Any of the detection probes described herein above or below may be used and similar labeling procedures can also be applied to the identification of differential matched ligand-protein binding pairs from multiple, related sources. For example, a mixture of proteins from normal tissue and a mixture of proteins from diseased tissue can be differentially labeled with a different dye or fluorescent label (and the like) for each of the protein mixtures. After incubation with the solid supports and washing away unbound protein, the differential ligand-protein binding pairs, those that demonstrate selectivity, that is, are specific to one set of proteins, are detected and isolated automatically or manually on the basis of the particular label or detection probe. In another embodiment, proteins are labeled with a detection probe, which is an affinity probe (tag) such as biotin. After incubation and washing of the proteins and solid supports to remove unbound protein, solid support comprising test compounds bound with tagged, for example, biotinylated proteins may be detected, for example, using streptavidin complexed with a phosphatase or a peroxidase. After addition of a suitable phosphatase or peroxidase substrate, the ligand-protein binding complex is detected.

In another embodiment, proteins bound to ligands can be detected using radioactiv- ity, i.e. the detection probe may be a radioactive compound. The proteins may be labeled with said radioactive compound by any conventional method. For example, the organism or cell is fed with a radioactive amino acid that is incorporated into its proteins. After incubation of the radioactive proteins with the solid supports and washing, bound radioactive protein is detected by, for example, autoradiography, and the ligand-protein binding pairs are isolated.

In yet another embodiment, particular classes of proteins that bind to test compounds can be detected using specific probes, for example, a family-specific antibody in an immunoassay such as an ELISA assay. Treatment with a conjugated monoclonal antibody for a family of proteins after incubation and washing, for exam- pie, provides information about the expression of related proteins. Where the protein mixtures are obtained from related protein sources, for example, from diseased and normal tissue, the ligand libraries can be incubated separately with each set of proteins. After detection and identification of the ligand-protein binding pairs, an assessment of the expression of the particular protein class in each state (for example, normal vs. diseased) can be determined. A monoclonal antibody may be conjugated to a fluorescent dye or to an enzyme such as peroxidase or alkaline phosphatase for quantification by ELISA. The antibody may also be conjugated to ferromagnetic beads by known, routine techniques. The magnetic beads concentrate near the location of the protein forming a "rosette" around solid support beads, or on the membrane sheet, or thread for detection.

In an additional embodiment, for example, where a single protein mixture is used, the extracted protein may be immediately incubated with the solid supports, and, after washing, detection of bound protein and isolation of the specific ligand-protein binding pair is done without any labelling, for example by measuring refractive index changes of the resin beads. Beads containing both proteins and ligand will have a different refractive index than beads containing only ligand. The refractive index changes could be detected from the light scattering when using an automated bead sorter as described below; or by using a custom-made instrument based on the principles of surface plasmon resonance or (SPR).

It is, however, preferred that at least one protein mixture is labelled using a fluorescent label, it is even more preferred that all protein mixtures are labelled using different fluorescent labels.

Non-limiting examples of methods for preparing labelled protein mixtures, such as differentially labelled protein mixtures are described in examples 11 , 12, 13 and 46 of WO2004/062553.

Bound ligand-protein complexes or pairs can be isolated from the bulk of solid supports, such as resin beads by various means depending on the nature of the detection probe. Isolation methods include for example, manually sorting resin beads containing bound labeled protein by detecting the detection probe for example with the aid of a microscope or sorting by fluorescence or by color depending on the screen- ing process used. Alternatively, the sorting process may be automated with the use of a beads sorter, such as by use of "fluorescence activated beads sorting" (FABS), for example specially designed, commercially available bead sorters may be used (e.g. Union Biometrica, Sommerville, Mass.) and detecting fluorescence intensity (Meldal, 2002, Biopolymers, 66: 93-100). In general, resin beads can be sorted at a rate of about 100 to 200 beads per second, or even faster depending on the equipment used and its reading capacity. A range of about 5 to 500, such as 5 to 110, preferably about 5 to 50 beads per second is sorted with known instruments. Slower rates may be used to increase accuracy. Preferred, is a rate where reading for example, only one resin bead passes through the detector at a time.

Identification of compound

Once a resin bead has been selected, the compound of said bead may be identified. In addition, the protein member of the ligand-protein binding pair of a selected resin bead may also be identified. Preferably, only one resin bead is used at a time. Thus if said resin bead only comprises one library member in one or more copies, then only one compound is identified at a time.

The library member and optionally the protein and ligand binding partners may be identified using any conventional technique known to the person skilled in the art, for example any of the techniques described herein below. It is preferred that the selected resin bead is isolated and that identification of the ligand and optionally identification of the protein is carried out on the isolated resin bead. It is thus preferred that either the library member or the protein or even both are identified using "on- bead" mehods. "On-bead" refers to methods wherein the identification processes or part of each identification process is performed directly on a bead, for example methods wherein the library member and/or protein are identified on the bead by for example spectroscopy or to methods wherein the library member and/or protein is enzymatically digested directly on the bead. In a preferred embodiment of the invention, identification of protein and protein ligand binding partners is identification from the protein/ligand complex on same, single bead. In this embodiment, resin beads comprising polyethylene glycol, preferably PEG-based resins with a size in the range of 300 - 800 μm are preferably used. In one embodiment, a resin bead con- taining the binding pair is cut into two unequal portions. One portion of the bead is used to identify the library member, while the other portion may be used to identify the protein. In another embodiment, the protein in first broken down into its constitutive peptides enzymatically or chemically (vide infra) and the library member is then released. Both the library member and the protein peptides may then simultane- ously or sequentially be analysed by for example mass spectrometry (vide infra). In this embodiment, the library member may be first analysed by NMR (vide infra) before break down of the protein and release of the library member. In one particular embodiment, the library member may be linked to the solid support via a methionine residue and the protein and library member can be simultaneously broken down and released by treatment with CNBr.

In one embodiment of the invention resin beads comprising protein ligands are first selected and subsequently proteins are released from the resin beads and the resin beads are screened for compounds modulating a cellular response. In such an em- bodiment the protein will in general not be associated with the finally selected resin beads. Thus identification may involve only identification of the library member according to the methods outlined herein below. Preferably, however, the protein member of the ligand-protein binding pair may be identified already after the first part of the process, i.e. directly after isolation of solid supports comprising ligand- proteinn binding pairs. It is also possible that the protein of a ligand-protein binding pair may be eluted or otherwise released from the solid support and stored. Once solid supports comprising a compound modulating a cellular response have been identified the corresponding stored protein may be identified.

The process for identification of the library member depends on the type of library used. For example the library member may be identified using mass spectrometry, NMR spectroscopy, infrared (IR), elemental analysis or combinations of the aforementioned. For a library of primarily oligomeric compounds, the library member can be analysed by Mass Spectroscopy (MS), particularly if the library was synthesized in such a way that the synthetic history of the compound is captured, for example, using a capping procedure to generate fragments of the compound that differ in mass by one building block (see, for example, Youngquist et al., 1995, J. Am Chem. Soc, 117: 3900-06). This capping procedure is most efficient when the cap and the building block are reacted at the same time. The capping agent can be any class of compound that has at least one functional group in common with the building block used to generate the oligomer, so that both the capping agent and the building block can react when added to the resin in an appropriate ratio. Alternatively, the capping agent can have two functional groups in common with the building block where one of the groups in common, such as the group in the building block that is used for the elongation of the oligomer, is orthogonally protected. For example, in a synthesis of a peptide using the Fmoc strategy, the capping agent could be the same as the building block but with a Boc group protecting the reactive amine instead of the Fmoc group (see St. Hilaire et al., 1998, J. Am. Chem. Soc, 120: 13312-13320). In another example, if the building block is a protected haloamine, the capping agent could be the corresponding alkylhalide.

Where the library is synthesized by parallel synthesis (a parallel array), the compound can be identified simply by the knowledge of what specific reaction components were reacted in a particular compartment. The structure can be confirmed by cleavage of a small portion of compound from the solid support and analyzed using routine analytical chemistry methods such as infrared (IR), nuclear magnetic resonance (NMR) spectroscopy, mass spectrometry (MS), and elemental analysis. For a description of various analytical methods useful in combinatorial chemistry, see: Fitch, 1998-99, MoI. Divers., 4: 39-45; and Analytical Techniques in Combinatorial Chemistry, M. E. Swartz (Ed), 2000, Marcel Dekker: New York.

In a preferred embodiment however the library has been synthesised by a split-mix approach where the precise structure of the compound of a specific bead is unknown. In this embodiment, the library member can be identified using a variety of methods. The compound may be cleaved off the resin bead, and then analyzed using IR, MS, or NMR. If the library is attached to the resin bead by a cleavable linker, then the compound can be cleaved by cleaving said cleavable linker. For NMR analysis, larger beads containing approximately 5 nmoles of material are preferably used for the acquisition of 1 -dimensional (1-D) and 2-dimensional (2-D) NMR spectra. Furthermore, these spectra can be attained using high-resolution MAS

NMR (magic angle spinning nuclear magnetic resonance) techniques. Alternatively, high resolution-MAS NMR spectra can be acquired while the ligand is still bound to the solid support, as described for example, in Gotfredsen et al., 2000, J. Chem. Soc, Perkin Trans., 1 : 1167-71. The compound may also be identified by release of the compound and fragmentation by MS-MS in MALDI or electrospray mode.

Frequently, resin beads used for library synthesis contain about 100 to 500 pmoles of material, which is generally insufficient for direct analysis using NMR techniques. In such situations, the libraries can be synthesised with special encoding to facilitate identification of the library member. For a review of encoding strategies employed in combinatorial chemistry see: Barnes et al., 2000, Curr. Opin. Chem. Biol., 4: 346-50. Most coding strategies include the parallel synthesis of the encoding molecule (for example, DNA, PNA, or peptide) along with the library compounds. This strategy requires a well-planned, time consuming, orthogonal protecting group scheme. Fur- thermore, the encoding molecule itself can sometimes influence the cell leading to false positives. Alternatively, the library members can be encoded using radiofre- quency tags or using optical encoding, such as quantum dot encoding, spherical encoding or distance encoding. These methods alleviates the problem of false positives stemming from the coding tags, but is generally only useful for small libraries in a one-bead-one-compound system due to the sheer bulk of the radiofrequency tag. Alternatively, single beads can be analyzed in a non-destructive manner using infrared imaging. This method gives limited information and while useful for pre- screening, is not recommended for conclusive structural determination.

In a preferred embodiment of the invention the library member(s) comprised within selected resin beads are identified using mass spectrometry (MS). MS can be used alone to identify the library member. The library member can be cleaved from the resin bead, the molecular mass determined, and subsequently fragmented into subspecies to conclusively determine the structure. MS-based methods of compound identification are useful in this invention, as they require very little material, and can utilise pico- to femtomole amounts of compound. MS-based methods include for example ES-MSMS, MALDI-MSMS or ES-LC/MSMS.

The binding protein may be identified using any conventional method known to the person skilled in the art. For example, the protein may be extracted from beads and identified by for example gel electrophoresis, such as 1 D- or 2D gel electrophoresis, mass spectrometry, such as MALDI-TOF-MS or ES-MS, NMR, peptide sequencing, for example by Edman degradation, or any other suitable method. In one embodi- ment it is preferred that the protein is identified using "on-bead" methods (see herein above and below). In another embodiment it is preferred that the protein is identified after elution from the resin bead and separation on a 1 D or 2D gel.

In one embodiment, a resin bead containing a ligand-protein binding pair is cut into two portions. One portion of the bead is used to identify the ligand, while the other portion is used to identify the protein.

The protein may be identified, for example, by performing systematic degradation of the protein on-bead. Most often, the protein can be broken down into its constituent peptides enzymatically, for example using a protease, such as trypsin or other known peptidases. General protocols for enzymatic breakdown of proteins during proteomic analysis can be found, for example, in 2-D Proteome Analysis Protocols, AJ. Link (Ed), 1^st Ed, 1999, Humana Pr: Totowa. Given the hydrophilic nature of preferred resins, trypsin works efficiently on-bead, and can efficiently cleave native proteins as well as proteins that have been covalently modified with a detection probe. The number of reasonably sized peptides generated by enzymatic cleavage is improved if the proteins are first denatured. Denaturation is easily accomplished on-bead, for example, on PEG-based resins that are robust and solvated in most denaturants used, such as guanidine HCI and urea. Otherwise denaturation may be obtained by drastic changes in temperature and pH. Other cleavage enzymes may be used, for example, endoprotease Arg-C, endoprotease Lys-C, chymotrypsin, endoprotease Asp-N, and endoprotease GIu-C. Sometimes, chemicals such as CNBr (as described, for example, in Compagnini et al., 2001 , Proteomics, 1 : 967-74) and [c/s-Pd(en)(H20)₂]²⁺ (as described, for example, in Milovic et al., 2002, J. Am. Chem. Soa, 124: 4759-69) may be used to degrade a protein into its constituent peptides.

Alternatively, the identified library member can be resynthesized and coupled to an affinity support such as PEGA, sepharose or sephacryl, and the protein member purified by affinity chromatography. Unlabelled protein mixture is applied to the affinity column and, after washing of the unbound protein, bound protein is eluted with solubilized library member. This route is time and reagent consuming. The library member must first be synthesized and purified, and then attached to the affinity support. It should also be produced in sufficient quantities that the required concen- tration can be used to elute protein from the affinity column. Alternatively, buffers of different pH, high salt and/or denaturants can be used to elute protein. It can sometimes be difficult to elute multimeric proteins from affinity columns using a monovalent ligand because of avidity effects.

To expedite the process and alleviate the aforementioned problems, the protein can be degraded into peptides while still bound to its ligand-binding partner, and the generated peptides analyzed. For example, the protein ligand may be resynthesized on small scale (25-50 beads) on a useful resin, preferably the same resin used for library synthesis, such as PEGA4000 resin or PEGA6000 resin. After binding of unlabelled protein from the mixture and washing off the unbound protein, the ligand- protein complex can be immediately degraded into the constituent peptides either enzymatically or chemically, using known processes and reagents and the peptides analyzed, for example, by peptide mass fingerprinting, or other known methods. Using this process several ligand-protein complexes can rapidly be digested. This process can be readily automated. The protein bound to the protein ligand can be identified by any suitable method such as MS or Edman degradation sequencing. For general protocols on the identification of proteins using proteomics techniques, see, for example, 2-D Proteome Analysis Protocols, A.J. Link (Ed), 1^st Ed, 1999, Humana Pr: Totowa. Protein can be identified from its peptide mass fingerprint, for example, using the mass of some of the constituent peptides obtained from enzymatic digests. The mass of the mixture of peptides generated from the digested proteins can be determined using MALDI- TOF-MS or ES-MS. The peptide masses or fingerprints are used to search data- bases of known proteins and gene products to identify the protein(s). To increase accuracy of the protein identification in the absence of other limiting information such as pi and mass, the results of several digests using different processes for cleavage are combined. Instead of, or in addition to, generating peptide fingerprints, a single peptide from the protein can be fragmented, and its amino acid sequence determined. The sequence can be used to identify known and unknown proteins, for example, by comparing to protein databases. The use of MS to identify the pro- teins(s) is well suited to the degradation of protein complexes on single beads, since very little material is required for identification (pico - femtomole). Alternatively, proteins can be identified using N-terminal sequencing via Edman degradation; pro- vided that the N- terminus is not blocked. This generally requires larger quantities of material (picomole).

After identification of the compound it may be desirable to confirm the activity of said compounds by further in vitro and/or in vivo assays. For example, resin beads com- prising the identified compound and optionally an adhesion compound may be synthesized and the cellular response confirmed. In addition binding to the identified protein may be confirmed by in vitro binding studies. It is also possible to test identified compounds in in vitro assays in the absence of beads. Cells may for example be grown directly in a tissue culture dish, flask or coverglass and the identified com- pound can be added directly to the medium of said cells. If several reporter systems are available for the particular cellular response then preferably several different reporter assays may be tested in vitro, in order to identify very useful compounds. For example, induction of a signal transduction pathway by a G-protein coupled receptor frequently involves internalization of the G-protein coupled receptor as well as a transcriptional response. Reporter systems for both internalization and transcription may thus be tested.

Protein ligands

In one aspect the present invention also relates to the proteins ligands identified by the methods disclosed herein as well as to uses of these protein ligands.

Thus the present invention in one aspect relates to methods of modulating the activity of a cell surface molecule comprising the steps of a) Providing a protein ligand capable of modulating the activity of a cell surface molecule, wherein said protein ligand is identified by the methods of the invention b) Incubating said protein ligand together with cells expressing said cell surface molecule c) Thereby modulating the activity of said cell surface molecule

In another aspect the invention relates to methods of modulating the activity of cellular protein(s) comprising the steps of a) Providing a protein ligand capable of modulating the activity of cellular protein(s), wherein said protein ligand is identified by the methods of the invention b) Incubating said compound together with cells expressing said cellular protein c) Thereby modulating the activity of said cellular protein

In yet another aspect the invention relates to methods of modulating a signal transduction pathway comprising the steps of a) Providing a protein ligand capable of modulating a signal transduction pathway, wherein said protein ligand is identified by the methods according to the invention b) Incubating said compound together with cells expressing said cellular pathway c) Thereby modulating the signal transduction pathway.

In an additional aspect the present invention relates to methods of modulating the interaction between two or more cellular molecules comprising the steps of a) Providing a protein ligand capable of modulating the interaction between two or more cellular, wherein said protein ligand is identified by the methods according to the invention b) Incubating said compound together with cells expressing said two cellular molecules c) Thereby modulating the interaction between the two cellular molecules

In one embodiment the invention relates to compounds of the general structure

R₁ = Side chains of various natural and unnatural amino acids (3, 4, 12,

20, 21, 28, 31, 47, 64, 73, 104, 127)

R₂ = Various acyl groups (117, 118, 120, 121, 126)

R₃ = Various aryl and alkyl (108 - 112, 115-116)) n = 1 -3 (39, 40, 41)

The invention also relates to such compounds linked to a solid support. Accordingly, the invention also relates to compounds of the general structure:

Ri = Side chains of various natural and unnatural amino acids (3, 4, 12, 20, 21, 28, 31, 47, 64, 73, 104, 127)

R₂ = Various acyl groups (117, 118, 120, 121, 126) R₃ = Various aryl and alkyl (108 - 112, 115-116)) n = 1 -3 (39, 40, 41)

Spacer = /\^0

OH

.0

HMBA =

HO OH Above-mentioned numbers in () refer to compounds disclosed in Tables 1 , 2, 3, 7 and 9 of WO2004/062553.

Thus, Ri is preferably the side chain of an amino acid selected from the group consisting of amino acids 3, 4, 20, 21 , 28, 31 , 47, 64, 73, 104 and 127 as shown in Tables 1 , 2, 3, 7 and 9 of WO2004/062553.

R₂ is preferably any acyl group selected from the group consisting of the acyl groups 117, 118, 120, 121 and 126 as shown in Table 9 of WO2004/062553.

R₃ is preferably any aryl or alkyl group selected from the group consisting of the aryl and alkyl groups 108, 109, 110, 111 , 112, 115 and 116 as shown in Table 9 of WO2004/062553.

In a preferred embodiment of the invention the protein ligand is selected from the group consisting of compounds of the following structures:

Examples

Example 1

Preparation of resin beads with library and adhesion compound

General methods for solid phase peptide synthesis (SPPS)

General for chemical synthesis:

All chemicals described are commercially available and used without further purification. All solvents were HPLC-grade. PEGA - resins were purchased from VersaMa- trix A/S, Copenhagen. Each washing step lasted 2 min unless otherwise stated. Purifications were performed on a standard reverse phase HPLC using gradients of acetonitrile-water with various amounts of TFA.

Coupling of HMBA linker to PEGA - resin:

Dry PEGA - resin was swelled in DCM and washed with DMF (3x). 3.0 eq. HMBA, 2.9 eq. TBTU and 6 eq. NEM were mixed in appropriate DMF and allowed to react for 10 min. The mixture was added to resin and after 2h the resin was washed with DMF (6x), DCM (6x) and lyophilised.

General Procedure for Coupling of Amino Acid to HMBA-linker: Dry PEGA - resin with HMBA-linker was swelled in dry DCM. 3.0 eq. Fmoc- protected amino acid, 2.25 eq. MeIm and 3.0 eq. MSNT were mixed in appropriate amount of dry DCM and added to resin. After 1 h the resin was washed with DCM (3x) and the coupling was repeated as above once. After coupling for 1 h the resin was washed with DCM (6x), DMF (6x), DCM (6x) and lyophilised.

General SPPS Coupling Procedure:

The terminal amino acid on the resin was Fmoc-deprotected by treatment with 20% piperidine in DMF (1x2 min + 1x18 min) followed by washing with DMF (6x). 3.0 eq. Fmoc-protected amino acid, 2.9 eq. TBTU and 6.0 eq. NEM were mixed in appropriate amount of DMF and allowed to react for 10 min. The mixture was added to the resin and after 2h the resin was washed with DMF (6x). General Side Chain Deprotection Procedure:

Dry PEGA - resin with acid stable linker and compound and/or peptide was swelled in H₂O and the side chains was deprotected with 95% TFA (aq) (2x15 min). If Pmc groups were present cleavage time was 6h. The resin was washed with H₂O until washing water had pH = 5-7. The resin was then washed with DMF (1Ox), DCM (1Ox) and lyophilised.

General HMBA Cleavage Procedure: Dry PEGA - resin with HMBA linker and attached compound was swelled in water and NaOH (aq.) 0.1 M was added. After 2h HCI (aq.) 0.1 M was used for neutralisation and then AcN was added until the H₂O/AcN ratio was 1 :1 by volume. The resin was filtered off and the liquid was used direct for RP-HPLC or/and Q-TOF MS analysis if needed.

The above general procedures are used for solid phase peptide synthesis in the following unless otherwise specified.

Formation of a library linked via an internal amide nitrogen

Synthesis of FmocGly/AllocGly (ratio -1 :1) modified PEGA 1900 beads

PEGA 1900 beads (300-500 μm, 0.24 mmol NH₂/g) (4 g, 0.96 mmol ~ 1 eq)) were treated with a 1 :1 TBTLJ coupling mixture of FmocGlyOH (2 eq) and AllocGlyOH (2 eq) , NEM (17 eq) and TBTU (3.8 eq). Coupling time was 3.5 h. Beads were washed 10 x with DMF.

Coupling of aldehyde photolinker to the "core" Fmoc-Glycine.

FmocGly/AllocGly beads were standard Fmoc deprotected with 20% piperidine in DMF leaving the Alloc glycine untouched. Then standard TBTU coupling of the aldehyde photolinker (4-(4-formyl-2-methoxy-5-nitrophenoxy)butanoic acid) to the deprotected glycine (1 eq = 0.48 mmol). After coupling, the beads were washed with DMF and DCM and lyophilized.

Tafcι/e 7; 5 Phenylethylamines used for reductive amination of aldehyde linker

Table 2: Compounds used for BTC coupling

Table 3: Compounds used foracylation in wells 1-10.

Table 4: Compounds used for TBTU coupling

Table 5: Compounds used for TBTU coupling of amino acids and acylation with acyl chlorides

Library synthesis:

Step 1. Reductive amination of photo linker aldehyde:

5 portions of the above beads (0.8 g dry beads) were placed in 5 syringes of 5 mL.

The beads were swollen in DMF and phenylethylamines (see Table 1 ) were coupled to the free aldehyde on the photolinker by reductive amination as follows: 0.8 g dry beads is 96 μmols ~ 1 eq were pre-treated with the reaction solvent (DMF/HOAc/TEOF/EtOH, 1 :1 :1 :1). Then to each syringe was added one of the five phenylethylamines (20 eq) dissolved in the reaction solvent (400 μl_), and the the same solvent was added so the beads were covered. After 0.5 h NaBH₃CN (20 eq) was added and the beads stirred cautiously until all dissolved. After 1 h another portion of NaBH₃CN (20 eq) was added and the mixture left for additional 2 h. The beads were washed with DMF (1 Ox), DCM (10x), MeOH(10%HOAc) (2x), MeOH (1Ox), DMF(20% pip) (2x), and DMF (1Ox), DCM (1 Ox). The 5 samples were lyophi- lized over night.

Step 2. Mix and split of phenylethyl amines:

0.4 g of each of the 5 bead samples from above were swelled in DMF, mixed and transferred to a custom made 20-well library synthesizer such that each well con- tained approximate equal amounts of beads.

Step 3. BTC coupling of first amino acid:

Total amount of beads = 2.0 g ~ 0.24 mmol gives 12 μmol NH/well ~ 1 eq.

20 BTC-couplings from 11 different amino acids as shown in Table 2 was made as follows: Beads were pre-treated with a 1 :1 vol% THF/DIPEA solution for 5 minutes and drained. Of each amino acid in Table 2, 3 eq (36 μmol) was dissolved in dry THF (200 μl_) and BTC (1.67 eq) added as 200 μL of a freshly made stock solution in dry THF. Then 2,4,6-collidine (14 eq), as 200 μL of a freshly made stock solution in dry THF, was added and the resulting 20 suspensions left for 5 minutes. Each suspension was added to the respective well and after short mixing the synthesizer was sealed and left over night with gentle shaking. Next morning the beads were washed with THF (10x) and DMF (10x) without mixing the wells.

Step 4. Acylation of well 1-10: Beads in well 1-10 were washed with DCM (10x) and Boc deprotected with 30% TFA in DCM followed by wash with DCM (10x), DCM (5% DIPEA), DMF, and DCM (10x) . From each of the 10 acyl chlorides in Table 3 was made a solution of 10 eq acyl chloride (120 μmol) and DIPEA (20 eq) in dry DCM (400 μL) containing catalytic amounts of DMAP. The resulting solutions were added to well 1-10 and left with gentle shaking for 1 h. The reaction was repeated. After end reactions the beads were washed with DCM (1Ox), and DMF (1 Ox).

Step 5, Removal of Fmoc protection groups:

Well 1 -20 were standard Fmoc deprotected, followed by wash with DMF (1 Ox).

Step 6, Mix and split of the 20 wells

The content of the 20 wells was thoroughly mixed and re-distributed equally into the wells.

Step 7. Coupling of second amino acid (20 amino acids):

To each well was coupled an amino acid according to Table 4 by a standard TBTU coupling. Coupling time 5 h. After end reaction the beads were washed with DMF (10x).

Step 8, Coupling of third amino acid/acyl chloride (15 amino acids - 5 acyl chlorides):

The beads in all wells were standard Fmoc deprotected and for well 1 -15 standard TBTU coupled with the Boc- protected amino acids in Table 5. For wells 16-20 the beads were first washed with DCM (10x) and the N-terminal amines acylated with the five acyl chlorides listed in Table 5 (entry 16-20) analogous to step 4. After couplings all wells were washed with DMF (10x).

Step 9. Alloc deprotection of second "core" glycine:

The beads from the 20 wells were all combined in a 50 mL syringe and washed with CHCI₃ (5x) and with Ar-degassed CHCI₃ containing 5% HOAc and 2.5% NEM (5x). A solution of Pd(PPh₃)₄ (3 eq, 0.72 mmol) in Ar-degassed CHCI₃ containing 5% HOAc and 2.5% NEM (10 mL) was added to the beads and after bobling a few minutes with Ar the syringe was sealed with parafilm and left for 2 h. The beads were washed with CHCI₃ (10x), DMF (1 Ox)₁ DMF (5% DIPEA, 5% sodium diethyldithio- carbamate)(5x), MeOH (1 Ox), DMF DMF (2Ox). Kaiser test was positive.

Step 10, Attachment of HMBA linker: According to the standard procedure above.

Step 1 1. MSNT coupling of FmocGlvOH: According to the standard procedure above.

Step 12, TBTU coupling of FmocLvs(Fmoc)OH According to the standard procedure above.

Step 13. Attachment of adhesion peptides

Adhesion peptides were synthesized on the Fmoc-protected lysine such that the final beads had two adhesion peptides pr. library molecule. One batch of the library was attached adhesion peptide A, and a second batch with adhesion peptide B. Adhesion peptide A: (Boc-D-Ala-D-Arg(Pmc)-D-Lys(Boc)-D-Arg(Pmc)-D-lle-D- Arg(Pmc)-D-Gln(Trt)-Gly- :

Fmoc deprotection of lysine residues: According to the standard procedure.

Coupling: The adhesive peptide was synthesized directly (stepwise) on the library beads using the general SPPS coupling procedure. Alternative method: The purified peptide (Boc-D-Ala-D-Arg(Pmc)-D-Lys(Boc)-D- Arg(Pmc)-D-lle-D-Arg(Pmc)-D-Gln(Trt)-L-Gly-OH) (3 eq) was coupled to the lysine NH₂ groups using the general SPPS coupling procedure.

Adhesion peptide B: (Fmoc-D-Arg(Pmc)-D-Gln(Trt)-D-Arg(Pmc)-D-lle-D-Arg(Pmc)- : Analogous to synthesis of adhesion peptide A.

Deprotection of adhesion and library peptides

Final deprotection of protecting groups was performed according to the general procedure described above yielding the TCOB libraries ready for testing.

Example 2: Cells on beads

Cell attachment to beads Cell lines

Cell lines are attached to beads using the procedure described in example 3

Colon epithelial cells Preparation of epithelial cells from patient biopsy

• A length of colon is cut out by the surgeon at the hospital from a patient suffering from colon cancer

• Immediately hereafter the colon is cut open and a biopsy is taken from the tumor (away from the tumor centre) as well as from the healthy tissue (as far away from the tumor as possible).

• The biopsies are placed in each their 50ml tube with ice cold Thyrodes buffer (NaCI 135mM, KCI 4mM, MgCl2 1 mM, NaH2PO4 0.33mM, Hepes 1OmM, Glucose 0.16%, CaCI2 25uM, BSA 0.05%, penicillin 333U/ml, Streptomycin 0.30mg/ml, Gentamycin 0.16mg/ml)

• Preparation of the biopsies is started within 1 hrs after cut out:

Tumor biopsy: o The tumor biopsy is transferred to a 5 cm sterile petri dish and kept wet with Thyrodes buffer during the preparation

o The tumor is cut into pieces of 1 x1 mm

o The process is continued as described under "Tumor and normal biopsies"

Normal tissue biopsy: o The normal biopsy is transferred to a 5 cm sterile petri dish and kept wet with Thyrodes buffer during the preparation

o The normal biopsy is scraped with a scalpel to separate cell tissue from mucous

o The process is continued as described under "Tumor and normal biopsies"

Tumor and normal bioipsy o The tumor pieces or normal cell tissue is transferred to a 50ml sterile tube with 20ml Thyrodes buffer + collagenase 1.5mg/ml (collagenase Worthington type II) + EDTA 0.0016% and is incubated for 10min at 37 degrees and mixed every 3 min o Supernatant (=supernatant 1 ) is transferred to two 10mI tubes (sterile) and centrifuged for 3 min at 400 rpm.

To the pellet (=pellet 1) is added another 20ml Thyrodes buffer + col- lagenase 1.5mg/ml (collagenase Worthington type II) + EDTA

0.0016% 30ml and it is incubated as described above

o After centrifugation of supernatant 1 the new supernatant is gently removed and saved for re-use Pellet 1 (cells) is re-suspended in 1 ml Growth medium (DMEM w.Glucose 4.5g/l, FCS 10%, penicillin 100u/ml, Streptomycin O.i mg/ml, Gentamycin 0.05mg/ml)

o The procedure is repeated 3 times re-using the Thyrodes buffer + col- lagenase 1.5mg/ml (collagenase Worthington type II) + EDTA

0.0016%30ml giving 3x 1 ml cell suspension

o The 3 cell suspensions from tumor biopsy consisting of various colon cell types are mixed and so are the 3 cell suspensions from normal colon biopsy and epithelial cells are isolated using Dynal bead separation.

Dvnal bead separation

CELLection™ Epithelial Enrich Dynabeads from Dynalbiotech are used to separate epithelial cells from other colon cells. The beads are superparamagnetic polystyrene beads coated with a monoclonal mouse IgGI antibody (Ber-EP4) via a DNA linker to provide a cleavage site for cell release. Ber-EP4 is an anti-EpCAM (Epithelial Cell Adhesion Molecule) specific for two glycopolypeptide membrane antigens expressed on most normal and neoplastic human epithelial cells.

Standard procedure provided with the CELLection™ Epithelial Enrich Dynabeads are followed

Separated epithelial cells from the two biopsies are re-suspended in 5ml growth medium (see above) and seeded in a T25 cell culture flask After 2 days culture cells are loosened using EDTA and seeded on control beads and library beads. Colon epithelial cells are attached to beads using procedure described in example 3

Myocyte attachment to beads

Cardiac myocytes is an example of another primary cell type that has been isolated and attached to the resin beads. The myocytes were prepared from 1 to 5 day old neonatal Wistar rats (University of Copenhagen) according to literature procedure described in Busk et al., 2002, Cardiovasc. Res., 56: 64-75 and plated into eight P10 culture plates at 6 million cells/plate. Cells were grown at 37⁰C and 5% CO₂ humidity in serum free Modified Eagle Media (MEM). After 2 days, the cells were attached to beads using the procedure described in example 3. The adherent cells were washed at room temperature with serum free MEM (2x) and fresh MEM was added.

Figure 2 shows primary rat myocytes on PEGA resin beads.

Alternative cell attachment to beads:

Cells can be attached to beads in a plate format using following procedure:

• Add beads to the wells of a 96 well, 24 well or other plate for- mat.

• Add cell suspension 5x10E6 cells/ml to the wells (volume: 96wΘll: 200ul, 24well 1 ml)

• Incubate o/n at 37 degrees, 5% CO2

Example 3: Cell functional screening

Partial release of compound:

The library compounds are partially released from the beads by UV exposure. It has been shown that the amount of released compound is proportional to the UV exposure time. It has furthermore been shown that release of an amount of compound sufficient to give a functional response leaves sufficient compound on the bead for the subsequent structure elucidation conducted after a second UV release.

The functional response is obtained by measuring apoptosis, which in the present assay is relying on membrane integrity.

- Membrane integrity assay:

- U2OS cells are transfected with eGFP and held under selection with G418 for 3 weeks to obtain a cell line stably expressing GFP

- U2OS-GFP cells are seeded on resin beads prepared as described in Ex- ample 1. The cells are seeded on Negative control beads (AC-NH-PLL-GIy-

Bead-Gly-HMBA-Gly-Lys-(NH-AP10)x2 and Positive control beads (2CX2- PPL-Gly-Bead-Gly-HMBA-Gly-Lys-(AP10)x2

- Add 1000 beads to a 1.5ml Eppendorf tube

- 1 ml cell suspension 0.5x10E6/ml Hams w. 5% FCS is added - Leave tube vertically in incubator (37 degrees, 5% CO2) for 16-24hrs - rock tube gently every 15min for the first hour

- Growth medium is removed and cells are incubated with 1 uM Ethidium Bromide in Hams F12 w. FCS 10%, penicillin 100u/ml, Streptomycin 0.1 mg/ml for 30min at 37 degrees. - Medium is replaced with growth medium + Acid red 5OuM.

- Beads are moved to a Nunc 24 well plate and placed on a Zeiss Axiovert 200M microscope for image acquisition. Images are acquired using Green filter setting (Excitation 480/30, Emission: 535/40, Dichroic: 505/DCLP) and Red filter settings (Excitation: 540/25, Emission: 605/55, Dichroic: 565DCLP). A positive response is characterized by a high Red/green ratio as a result of membrane permeability in apoptotic cells giving efflux of GFP and influx of Ethidium Bromide.

- Alternatively beads can be sorted using a Fluorescence Activated Bead Sorter where beads are sorted based on laser excited red/green fluores- cence

Negative and positive beads are mixed in the functional assay described above and compound is partial released from the beads by UV exposure (400W) for 60 sec. Beads are analysed using microscope platform as described. Results are shown in fig. 3. Beads showing a positive response are transferred to proteome screening as described in example 5

Example 4

Clearance of cells and adhesion peptide

Library beads that have been screened as described in Example 3 and isolated after giving a positive response as described in Example 3 and with cells still attached were drained from medium and washed with water (2x) in a syringe equipped with a teflon filter. The beads were treated with 0.1 M NaOH solution for 1 h then washed with water (5x), DMF (5x), DCM (5x), acetonitrile (1Ox), DMF (1Ox ) then water (2Ox) thereby removing attached cells and adhesion peptide.

Example 5

Proteome screening of active compounds

Colon cancer cells/tissue are homogenised and lysed in IP-buffer (IP buffer is 50 mM HEPES (pH 7.5); 10% glycerol; 150 mM NaCI; 0.1 % Tween-20 to which protease inhibitors are added by dissolving one tablet of Complete Protease Inhibitor Cocktail (Complete Mini; Roche; cat.#: 1836153) in 10 ml IP-buffer) and followed by immediate snap freezing in liquid nitrogen. After thawing, the lysate is sonicated and centrifuged to remove insoluble cellular components. The supernatant is collected and the protein concentration determined by using Bio-Rad Protein Assay Dye Reagent from Biorad (cat.#: 500-0006).

Lysate (50μl; 5 μg/μl) is combined with active beads from Example 4 in one micro- tube and with control beads in another microtube and incubated overnight at 4⁰C. Both sets of beads are then washed 4 times with ice-cold IP-buffer and once with 2x with ice-cold water.

The active beads are placed one each into individual microtubes and protein bound to the beads is identified by mass spectrometry as described below. Control beads are processed similarly: The bead is washed with 15 μl 100 % acetonitrile then placed in a speedvac until completely dry. The dry bead is mixed with 15 μL DTT (10 mM in 0.1 M ammonium bicarbonate) at 56°C for 1 hour. After cooling, the DTT is removed and 50 mM io- doacetamide in 0.1 M ammonium bicarbonate (15 μL) is added. The mixture is incubated in the dark for 30 minutes at room temperature. DTT (50 mM in ammonium bicarbonate (ca. ¹Λ volume) is added to react with excess iodoacetamide. Trypsin in 50 mM ammonium bicarbonate buffer (12.5 ng/μL) is added and the sample incubated at 37⁰C overnight. The protein peptides are extracted as follows: Transfer the supernatant to an eppendorf tube Add 15-20 μL 25 mM NH₄HCO₃ and mix for 10 minutes Add 25-30 μL 100 % acetonitrile and mix for 10 minutes Remove the liquid and pool it with the supernatant Add 15-20 μL 5% HCOOH and mix for 5 minutes Add 10 μL H₂O and mix 5 minutes Add 20 μL acetonitrile and mix 5 minutes Remove the liquid and pool it with the supernatant Evaporate the liquid from the pooled fractions in a speed vac to ca 2-6 μL.

The peptides are micropurified on reverse phase resin and after elution, the peptides (0.5 μl) are applied to a to a stainless steel disc to which 0.5 μL of CHC matrix +1.0 % TFA is added. The remaining solution is transferred to a new tube and stored at -2O⁰C.

Mass spectra are acquired on MALDI mass spectrometer operated in the positive reflectron mode using delayed extraction. Both MALDI-TOF and MALDI-TOF-TOF are used. The spectra are calibrated using bradykinin peptide or in some cases, internal mass calibration is performed using trypsin autolysis peaks or keratin peaks. Using MALDI-TOF, peptide peak masses are searched against peptide mass maps in different protein databases, for example NCBI or Swiss-Prot, using the following search engines found on the world wide web (www.) for each of: MS-FIT (prospector.ucsf.edu/ucsfhtml/msfit. htm), Profound (129.85.19.192/profound_bin/WebProFound.exe), and

MASCOT (matrixscience.com).

A molecular mass range is estimated from 0-250 K Da, allowing a mass accuracy that varied from 0.1 Da (some cases 0.3 Da) for each peptide mass. A large pi range from 0-14 or 0-12 is considered for each search. If no proteins match, the mass window is extended. Partial enzyme cleavages allowing for two missed cleavage sites and modification of cysteine by alkylation are considered in the search approaches. A protein is considered identified if the matched peptides cover at least 30 % of the complete sequence. A match of less than 30 % is considered in some cases, if prominent peaks are obtained. Usually, four or more peptides are used for identification. In some instances, hypothetical proteins or gene products, such as biochemical material, either RNA or protein, calculated from the expected expression of a gene and to which a function may be assigned based on sequence homology, are identified. When using MALDI-TOF-TOF, the individual peptide peaks from the MALDI-TOF spectra are further fragmented into their individual amino acids and the spectra generated used to search for matches to a theoretical fragment spectrum in a sequence database.

Example 5a

The specific protein binding to resin beads comprising compounds eliciting the cellular response of apoptosis identified as described in Example 3 is identied using the protein identification methods described in Example 5. The protein is X-linked Inhibitor of Apoptosis (XIAP); gi 1 184320.

Example 6

Identification of active library compounds from single bead

Individual beads from of Example 5 are transferred to separate microtubes and washed extensively with water to remove all peptide residue. They are then swelled in water and irradiated for 20 min with an OMNILUX E-40 (400W UV lamp, 365 nm, # 89514005, Steinigke Showtechnic GmbH, Germany). Acetonitrile:water (1 :1 ) 40 μL is added and the beads spun to the bottom of the tube by centrifugation. The supernatants are removed analysed by MS on an Bruker Esquire 3000 ion-trap mass spectrometer. The compounds are fragmented and their identities are deter- mined based on the masses of the fragments generated. Example 6a

By use of the methods described in Example 6, two beads that had elicited a positive response after screening as described in Example 3 and had shown binding to the protein XIAP were identified as having the following structures:

Binding to XIAP was demonstrated by incubation of resin beads with a proteome and subsequent Western blotting for XIAP.

Example 7

Proteome screening for active compounds for colon cancer

A small molecule combinatorial library synthesized as described in Example 36 in WO2004/062553 was used for binding proteins that are differentially expressed in colon cancer and consequently are putative protein targets for the development of anti cancer drugs. The proteome screening process was carreied out using protein mixtures that were differentially labelled with a fluorescent detection probe as described below:

Normal and cancer biopsies from a colon cancer patient were first rinsed of DMEM transport media then dissected into smaller pieces (ca. 2 x 2 mm). The biopsies were then metabolically labelled overnight with [³⁵S]-methionine in a methionine-free DMEN medium in a 37 ⁰C humidified incubator supplemented with 5% CO₂. Protein was extracted from the biopsies by homogenisation in 200 μL of Homogenisation Buffer [10 mM Phosphate Buffer, pH 7.2 augmented with 137 mM NaCI, 50 mM KCI, 5 mM CaCI₂, 5 mM MgCI₂, 8 M urea, 2% CHAPS and 30 mM DTT]. An additional

200 μL of homogenisation buffer and a DNAse/RNAse mixture (Amersham) were added and the samples incubated for 30 min at 4⁰C.

The suspensions were centrifuged and the supematants removed. The super- natants were dialysed 1x against Screening Buffer [10 mM Phosphate Buffer pH 7.2 augmented with 137 mM NaCI, 1 mM CaCI₂, 1 mM MgCI₂, 1 mM MnCI₂, 1 mM ZnCI₂ and 0.5 mM CuSO₄] for 1 hour @ 4⁰C.

The respective supematants were then labelled with Cy2 (minimal labelling dye)(Amersham Biosciences now GE Healthcare) and Cy3 (minimal labelling dye) (Amersham Biosciences now GE Healthcare), dyes according to the manufacturer's protocol. Normal protein extract was labelled with Cy3 and cancer was labelled with

Cy2. After labelling, the samples were dialysed using Amershams Ettan mini dialysis kit (1000 Da mwco) against screening buffer for 3x 1 hour at 4⁰C.

Portions of the labelled extracts were immediately frozen in dry ice for 2D-gel elec- trophoresis, while the other portions were mixed in equal quantities and used for screening the library.

Library screening: Prior to incubation, the library beads were washed with di- chloromethane (1 x 5 min), acetonitrile (1 x 5 min), methanol (1 x 10 min), water (2 x 10 min) and screening buffer (3 x 10 min).

The protein mixture was added to the beads in a syringe fitted with a teflon filter and incubated overnight (ca. 10 h) at 4⁰C.

The excess protein was removed by suction, and the beads washed with screening buffer (4 x 15 min) and water (3 x 20 min).

Sorting: The same day, the library was then sorted using a COPAS Biosort FABS sorter (From union Biometrica). The beads were added in batches to sheath fluid and ran through the sorter selecting beads with a high green fluorescence. The non-selected beads were then resorted selecting those with a high red fluorescence. The beads with a high green fluorescence were then resorted selecting those with a high green but low/no red fluorescence. And likewise the beads with high red fluorescence were resorted selecting those with a high red but low/no green fluorescence. After sorting, the beads were stored at 4⁰C. The next day, the sorted beads and remaining library members were examined un- der a fluorescence microscope to check accuracy of sorting and to qualitatively rank the beads based on fluorescence intensity. Selected beads i.e. beads containing mainly fluorescence from cancer proteins were then transferred Eppendori tubes, 65 μL of electrophoresis buffer was added and the sample immediately frozen. 2D GeI Electrophoresis: The protein(s) bound to individual positive beads were extracted into the electrophoresis buffer and were separated using standard 2D gel techniques using IPG of 4-7 and 6-9 gel strips.

After this step resin beads are ready to be subjected to screening for resin beads comprising compounds capable of modulating a cellular response. Thus, cells may be added to the resin beads after this step (see details herein below).

The total protein extracted from the biopsies was also separated on a 2D Gel. After separation, the gels were dried on Whatmann 3MM paper and exposed to phospho- imager plates for 10-30 days. The gel images were analysed using an in-house pattern analysis program and protein spots that were up-regulated in the cancer samples were identified. Since there was too little protein from the beads for unambiguous identification by MS, the corresponding protein in the gel of the total protein extract was located since it was present in larger amounts. The protein spots were cut out of the gel and the protein identified by mass spectrometry using standard proteomics techniques. By use of the procedures described here in this Example 7, a protein specifically overexpressed in human colon cancer tissue and binding to a specific protein ligand of a combinatorial library prepared as described in Example 36 of WO2004/062553 was identified as vinculin (metavinculin); swiss prot P18206.

After identification of resin beads with high red/low green or high green/low red fluorescence each of the fluorescent beads may be separated into individual wells of micro titre plates. The fluorescently labeled protein is eluted of the beads and the structure of the eluted protein(s) can be identified using MS as described above.

After removal of the attached protein, cells are allowed to attach the each of the beads using the procedure described in example 3. A fraction of the library compound is released by cleaving a photo-labile linker by illuminating with an OMNILUX E-40 UV light (365 nm) for 30 sec (as described in example 3). The effect of the released compound is tested on either of three different cell functional assays: a) induction of apoptosis (as described in example 3); b) effect on migration (by using the InnoCyte Cell Migration Assay Cat. No. CBA010 from Calbiochem) or c) changes in the morphology of the actin cytoskeleton. The latter assay is conducted by staining actin with phalloidin.

Example 8

Identification of compounds binding to overexpressed cancer proteins.

A resin bead identified by the proteome screening described in Example 7 from which the bound protein had been removed and identified as described in Example 7, was washed in a microtube with water (5 x). 5 uL of 0.5% triethyiamine/H₂O was added to the bead for 30 min to release the compound from the bead. Acetonitrile (200 uL) was added. The sample is centrifuged and the supernatant was collected. Electrospray Mass spectra of the sample was on a Bruker-Daltonics Esquire 3000 Plus instrument. The sample was injected at 180 uL/hour. A spectrum of control sample was run to determine all the irrelevant signals coming from, for example, residual buffer compounds. The Molecular ion was detected and confirmed in both positive and negative mode. The molecular ion was fragmented and spectra of the generated ions collected in both modes (See figure 4). A computer program that calculated the masses of all possible compounds, by-products and fragments was tailor-made for analysing this library. By iterative combination of the masses from the computer program and the molecular ion and fragment masses from the mass spectra, it was possible to identify the compound which bound the proteins. By application of the procedures described in this Example 8, the ligand binding the protein vinculin identified as described in Example 7, has the structure shown below.

Abbreviations:

HGF: Hepatocyte Growth Factor

NGF: Nerve Growth Factor PDGF: Platelet Derived Growth Factor

FGF: Fibroblast Growth Factor

EGF: epidermal Growth Factor

GH: Growth hormone

TRE: TPA Response Element SRE: serum response element

CRE: cAMP response element

AcN: acetonitril;

Boc: tert-butoxycarbonyl; tBu: tert-butyl; DCM: dichloromethane;

DMF: dimethylformamide;

Fmoc: 9-fluorenylmethoxycarbonyl;

HMBA: 4-hydroxymethylbenzoic acid;

Q-TOF MS: quadrupole time-of-flight mass spectrometry; MeIm: N-methyl imidazole;

MSNT: 1 -(mesitylene-2-sulphonyl)-3-nitro-1 H-1 ,2,4-triazole;

NEM: 4-ethyl morpholine;

PEGA: polyethylene glycol-polydimethyl acrylamide resin;

Pf p : pentafluorophenyl; Pmc: 2,2,5,7,8-pentamethylchroman-6-sulfonyl;

RP-HPLC: reversed phase high pressure liquid chromatography;

SPPS: solid phase peptide synthesis;

TBTU: 0-(benzotriazol-1-yl-N,N,N',N'-tetramethyluronium tetrafluoroborate;

TFA: trifluoroacetic acid; Thi: thienyl

Fur: furanyl

BzThi: benzothienyl

Claims

1. A method of identifying a protein ligand modifying at least one cellular response, said method comprising the steps of:

(a) Providing multiple solid supports capable of supporting growth of cells, wherein each solid support is covalently linked to one member of a library of test compounds and wherein at least two solid supports comprise different library members; and (b) Attaching cells onto said solid support, wherein said cells comprise reporter system^) for each cellular response, wherein said reporter system(s) generate detectable outputs or can be linked to a detectable output; and

(c) Screening said solid supports for solid supports comprising cells meeting at least one predetermined selection criterion, wherein said selection criterion is linked di- rectly or indirectly to said detectable output; and

(d) Selecting solid supports comprising cells meeting said at least one selection criterion, thereby obtaining selected solid supports; and

(e) Releasing cells from the selected solid supports by cleaving the cleavable linker; and (f) Providing one or more protein mixture(s); and

(g) Incubating the selected solid supports with the one or more protein mixtures; and

(h) Detecting ligand-protein binding pairs;

(i) Isolating solid supports comprising iigand-protein binding pairs; and

0) Identifying said the library member of the ligand-protein binding pair, thereby identifying a protein ligand modifying said at least one cellular response,

wherein the steps of the method may be performed in any suitable order.

2. The method according to claim 1, wherein the steps of the method are performed in the order outlined in claim 1.

3. The method according to claim 1, wherein the steps of the method are performed in the following order: (a), (f), (g), (h), (i), (b), (c), (d), (e) and G).

4. The method according to claim 1, wherein the method furthermore comprises the step of identifying the protein member of the ligand-protein binding pair.

5. The method according to any of the preceeding claims, wherein the one or more protein mixtures are labelled with a detection probe.

6. The method according to any of the preceeding claims, wherein at least two protein mixtures and wherein one protein mixture is labelled with one detection probe and another protein mixture is labelled with another detection probe.

7. The method according to any of the preceeding claims 1 , wherein the solid supports are resin beads.

8. The method according to claim 7, wherein the resin beads are selected from the group consisting of Toyopearl, sepharose, sephadex, CPG, silica, POPOP, PEGA,

SPOCC, Expansin, Tentagel, Argogel, Polystyrene, Jandagel, polydimethylacryla- mide resin, Polyacrylarnide resin, kieselghur supported resins and polystyrene supported resins.

9. The method according to any of the preceeding claims, wherein the solid supports are linked to an adhesion compound.

10. The method according to claim 9, wherein the adhesion compound is linked to the solid support via a cleavable linker.

11. The method according to any of the preceeding claims, wherein the cell adhesion compound is a peptide with an overall positive netcharge.

12. The method according to any of the preceeding claims, wherein said cellular response is modulation of a signal transduction pathway

13. The method according to any of claims 1 to 12, wherein said cellular response is modulations of signal transduction pathway, wherein said modulation is selected from the group consisting of • Upregulation or downregulation of the level of a member of the pathway;

• Relocalisation of a member of the pathway;

• Complex formation between members of the pathway or between members of the pathway with other cellular compounds; • Enhanced or reduced transcription from genes regulated by the pathway;

• Modification by for example phosphorylation of a member of the pathway;

• Activation or inhibition of an enzyme of the pathway;

• Degradation of a cellular compounds due to upregulation or downregulation of the pathway; • Altered secretion of a compound;

• Change in ion-flux;

• Morphological changes; and

• Change in viability

14. The method according to any of the preceeding claims, wherein said cellular response is modulation of a signal transduction pathway mediated by a cell surface molecule.

15. The method according to claim 14, wherein said cell surface molecule is a G- protein coupled receptor (GPCR).

16. The method according to claim 14, wherein said cell surface molecule is a receptor selected from the group consisting of receptors belonging to the family of protein kinase coupled receptors (e.g. cytokines, interferons, HGF) and receptors belonging to the family of receptor kinases (e.g. Insulin, NGF, PDGF, FGF, EGF, GH)

17. The method according to any of claims 1 to 11, wherein the cellular response is modulation of transcriptional activity.

18. The method according to claim 17, wherein said transcriptional activity is regulated by a response element.

19. The method according to any of claims 1 to 11, wherein the cellular response is change in the intracellular level of a compound

20. The method according to claim 19, wherein said compound is Ca²⁺

21. The method according to claim 19, wherein said compound is cAMP

22. The method according to any of claims 1 to 11, wherein the cellular response is relocalisation of a compound.

23. The method according to claim 22, wherein said relocalisation is relocalisation of a cell surface receptor from the cellular membrane to the cytoplasma.

24. The method according to any of claims 1 to 11, wherein the cellular response is change in pH in an intracellular compartment.

25. The method according to any of claims 1 to 11, wherein the cellular response is a change in a membrane potential.

26. The method according to any of claims 1 to 11, wherein the cellular response is change in interaction between two or more cellular molecules.

27. The method according to any of claims 1 to 11, wherein the cellular response is formation of a complex between two or more cellular molecules.

28. The method according to any of claims 1 to 11 , wherein the cellular response is disruption of a complex between two or more cellular molecules.

29. The method according to any of claims 26 to 28, wherein the cellular response is change in interaction between proteins involved in regulation of apoptosis.

30. The method according to any of claims 1 to 11, wherein the cellular response is apoptosis.

31. The method according to any of the preceding claims, wherein the reporter system is a system endogenous to said cells.

32. The method according to claim 31, wherein the reporter system comprises the intracellular level of an endogenous compound.

33. The method according to claim 32, wherein said compound is Ca2+

34. The method according to claim 31, wherein the reporter system comprises the intracellular localisation of an endogenous compound.

35. The method according to any of claims 1 to 30, wherein the reporter system comprises a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to a response element, the activity of which is modu- lated by the cellular response.

36. The method according to claim 35, wherein the reporter system comprises a nucleic acid comprising a nucleotide sequence encoding a detectable polypeptide operably linked to a response element, the activity of which is modulated by said signal transduction pathway.

37. The method according to any of claims 1 to 30, wherein the reporter system is a selected from the group consisting of BRET reporter systems, FRET reporter systems and proximity ligation reporter systems.

38. The method according to any of the preceeding claims, wherein the detectable output is bioluminiscense.

39. The method according to any of the preceeding claims 1 to 37, wherein the de- tectable output is fluorescence.

40. The method according to claim 38, wherein one predetermined selection criteria is a quantitative level of said bioluminiscence above or below a specific threshold.

41. The method according to claim 39, wherein the predetermined selection criteria is specific localisation of a fluorescent signal.

42. The method according to any of the preceding claims, wherein said cells are selected from the group consisting of mammalian cells.

43. The method according to claim 42, wherein the cells are primary mammalian cells.

44. The method according to any of the preceding claims, wherein at least 100 resin beads comprising different library members are provided.

45. The method according to any of the preceding claims, wherein each resin bead does not comprise more than one library member in one or more copies.

46. The method according to any of the preceding claims, wherein the library is selected from the group consisting of libraries of natural oligomers such as peptides, glycopeptides, lipopeptides, nucleic acids (DNA or RNA), or oligosaccharides; unnatural oligomers such as chemically modified peptides, glycopeptides, nucleic ac- ids (DNA or RNA) or oligosaccharides; and small organic molecules.

47. The method according to any of claims 1 to 45, wherein the library is a library of small organic molecules.

48. The method according to any of the preceding claims, wherein protein ligands modifying at least two cellular responses are identified, wherein step c) involves screening said resin beads for beads comprising cells meeting at least two predetermined selection criterion, wherein each selection criterion is related to a different detectable output.

49. The method according to any of the preceding claims, wherein the library member is identified using mass spectrometry.

50. The method according to any of the preceding claims, wherein the library mem- ber is identified using NMR spectroscopy.

51. The process according to claim 4, wherein the protein is identified using mass spectrometry

52. A method of manufacturing a protein ligand modifying at least one cellular response, wherein said method comprises the steps of:

a) Identifying a protein ligand by the method according to any of claims 1 to 51 b) Preparing said protein ligand by chemical synthesis c) Thereby manufacturing said compound

53. A method of modulating the activity of a cell surface molecule comprising the steps of a) Providing a compound identified by the method according to any of claims 14 to 16 b) Incubating said compound together with cells expressing said cell surface molecule c) Thereby modulating the activity of said cell surface molecule

54. A method of modulating the activity of cellular protein(s) comprising the steps of a) Providing a compound identified by the method according to any of claims 1 to 51 b) Incubating said compound together with cells expressing said cellular protein c) Thereby modulating the activity of said cellular protein

55. A method of modulating a signal transduction pathway comprising the steps of a) Providing a compound identified by the method according to any of claims 12 to 16 b) Incubating said compound together with cells expressing said cellular pathway c) Thereby modulating the signal transduction pathway.

56. A method of modulating the interaction between two or more cellular molecules comprising the steps of a) Identifying a compound by the method according to any of claims 26 to 28 b) Incubating said compound together with cells expressing said two cellular molecules c) Thereby modulating the interaction between the two cellular molecules

57. Compound identified by the method according to any of claims 1 to 51.

58. A compound of the structure

R₁ = Side chains of various natural and unnatural amino acids (3, 4, 12,

20, 21, 28, 31 , 47, 64, 73, 104, 127)

R₂ = Various acyl groups (117, 118, 120, 121 , 126)

R₃ = Various aryl and alkyl (108 - 112, 115-116)) n = 1-3 (39, 40, 41)

wherein the numbers 3, 4, 12, 20, 21, 28, 31, 47, 64, 73, 104, 108-112, 115-116, 117, 118, 120, 121, 126 and 127 refer to compounds disclosed in Tables 1, 2, 3, 7 and 9 of WO2004/062553.

59. The compound according to claim 58, wherein the compound is immobilised on a solid support.

60. The compound according to claim 58, wherein the compound is of the structure

61. The compound according to claim 58, wherein the compound is of the structure

62. The compound according to claim 58, wherein the compound is of the structure