WO2006133476A2 - Enzyme array - Google Patents

Abstract

The present invention relates to a protein array comprising at least one enzyme selected from the group consisting of lipase and esterase bound to a solid support, wherein the solid support is covered at least in part with a hydrogel matrix and the enzymes are bound to said matrix.

Description

Enzyme Array

The present invention relates to a protein array comprising a lipase or an esterase and uses of said array.

The emerging field of protein microarray technology has provided several solutions for the detection of DNA-protein, protein-protein and small molecule-protein interactions [1-3] . Especially the detection of protein-protein (as well as peptide- protein) and small molecule-protein interactions is becoming increasingly important, mainly because there is a strong need to identify novel therapeutic substances. Microarray technology may be the technology of choice to achieve this object because it allows to detect many interactions between a variety of molecules and a variety of proteins in a quick and accurate way (especially when used in high throughput screenings) . A further advantage of this technology is the fact that only small amounts of protein and molecules have to employed. This is important for those investigations where binding partners are not available in high amounts. Particularly interesting is the detection of interactions between small molecules and enzymes because many enzymes known are involved in crucial biochemical pathways and are known to play a key role in several diseases.

Since many proteins of interest exhibit enzymatic activity an increasing effort has been made to detect activity-based interactions of proteins and their substrates, pseudo-substrates and inhibitors on microarray chips . In principle there are two major strategies to detect these interactions: 1. proteins/en-_^ zymes are immobilized on a microarray and the array is subjected to a number of different molecules, 2. molecules are immobilized on an array and said array is subjected to proteins/enzymes. However, it has to be considered that the accessibility of the active site for potential binding partners as well as the cata- lytical activity of the protein/enzyme is not affected when proteins/enzymes are immobilized on a solid support. Furthermore, when immobilizing small molecules, which are binding or potentially binding to a protein/enzyme, it has to be taken into account that the molecule moiety which is responsible for the interaction has also to be accessible for the active site of the protein/enzyme .

It is known in the art that the immobilization of proteins on solid supports is not trivial since many immobilization methods lead to inactive immobilized proteins which consequently can not be used for binding studies. This problem is known especially for lipases and esterases [6].

The article of Veronese F M et al. (Farmaco 2001, 56: 541- 547) discloses the entrapment of pegylated lipases in a hydro- gel. According to said publication lipase is modified with PEG, mixed with polyvinyl alcohol and formed into hydrogel cylinders. The lipase in the matrix exhibited a low activity compared to the "free" lipase.

Martin B D et al. (Biomaterials 1998, 19: 69-76) describe a method for manufacturing hydrogels to be used, for instance, for enzyme immobilization employing lipases.

In Mohapatra S C et al. (Biotechn Bioeng 1999, 64: 213-220) a method for entrapping pegylated lipase into Ca-alginate gel beads is disclosed.

Basri M et al. (Appl Biochem Biotechn 1999, 81: 205-217) describe the immobilisation of lipases in hydrogels.

The covalent immobilisation of hydrolases, in particular of organophosphorous hydrolase, is disclosed in Andreopoulos F M et al. (Biotechn Bioeng 1999, 65: 579-588) .

Oskolkova O V et al. (Biochim Biophys Acta 2002, 1597: 60- 66) report of phosphonate inhibitors containing a polarity-sensitive fluorophore to monitor the polarity in the active site of enzymes, in particular of lipases.

Scholze H et al. (Analyt Biochem 1999, 276: 72-80) describe fluorescent inhibitors for the characterisation of lipases.

The WO 2002/063299 relates to a method for screening immobilized ligands with a reduced affinity to a binding partner.

In the WO 92/05788 phosphonates are disclosed which may be used for inhibiting lipases .

Chip-based screening of lipoytic enzymes would be useful since these proteins play key roles in lipid metabolism and energy homeostasis. They are also important biocatalysts in the chemical and pharmaceutical industry. Functional screening and characterization is thus essential for a better understanding of these potential drug targets as pertinent to lipid-associated diseases like obesity, atherosclerosis or stroke.

The importance of lipases and esterases in many diseases shows the need in the art to have biochips/microarrays with said enzymes bound thereto. Such chips would, for instance, enable the person skilled in the art to screen inhibitors or other binding partners for said enzymes. However, to date no microar- ray chip comprising a lipase or esterase is available in the art. Chen et al . reported, for instance, that hydrolytic enzymes can be maintained in active form on epoxide-activated slides . Nevertheless, the approach chosen by said work group was successful only when the enzymes were proteases but did not work with lipases [6] .

Therefore it is an object of the present invention to provide a protein array comprising catalytical active lipases/esterases, which can be used to detect the interactions between said enzymes and small molecules.

This object is achieved by a protein array comprising at least one enzyme selected from the group consisting of lipase and esterase on a solid support, wherein the enzymes on said solid support are covered at least in part with a hydrogel matrix.

It surprisingly turned out that a hydrogel matrix can successfully be employed to cover on an array of lipases and esterases bound to a solid support without affecting their catalytical activity and concerving its sensitivity.

According to the present invention all known lipases and esterases can be immobilized catalytically active to a solid support comprising a hydrogel matrix. The lipases and esterases may be isolated from natural sources or produced recombinantIy. Furthermore, it is also possible to use fusion proteins (e.g. one lipase/esterase moiety and one or more other protein/peptide moieties, like streptavidin, Fc regions of antibodies) or proteins chemically modified. The enzymes may be immobilized on the surface of the solid support, directly via functional groups or indirectly via functional groups bound to linkers, which are bound to said surface. Furthermore, the enzymes may also be immobilized to the hydrogel present on the solid support.

Although the present invention may be used for all esterases and lipases, the enzymes preferably used in an array according to the present invention are the lipases and esterases which are used on an industrial level or which are of interest in medical diagnosis, especially e.g., adipose triglyceride lipase (ATGL), hormone sensitive lipase (HSL) , monoglyceride lipasse (MGL) , hepatic lipase (HL) , lipoproteinlipase (LPL) , endothelial lipase (EL) , pancreatic lipase (PALI) and cholesterol esterase (CE) .

According to the present invention the terms "array" and "microarray" refer to an arrangement of elements, in particular of proteins/enzymes, like lipases and esterases, present on a solid support. If proteins are the elements of an array, said array is denominated "protein array" . Arrays or microarrays present on a solid support may also be called "chip" or "bio- chip" (e.g. microarray chip). An array of proteins/enzymes can be arranged on such a biochip in low-density (e.g. 2 to 50 spots per cm²) or high-density (e.g. 50 to 500 and even more spots per cm²) . The biochip according to the present invention is typically in geometrically-regular shape allowing a facilitated fabrication, handling, placement, reagent introduction, detection and storage. Of course it can also be irregular. An array is preferably configured in a row and column format, with a defined space between the single elements of the array. Such an arrangement allows an easier handling especially in high throughput applications, in particular when robotic delivery, masking or sampling of the reagents is employed. Furthermore, a well arranged array can also facilitate detection or quantitation by, for instance, scanning by laser illumination, confocal or deflective light gathering, CCD detection and chemical luminescence.

"Solid support", as used herein, refers to any solid support that can be employed in the invention and comprises or is composed of, e.g. glass, silica, modified silicon, ceramic, plastic, ZrO₂, Gold or any type of appropriate polymer such as polystyrene, polyethylene, polypropylene, (poly) tetrafluoro- ethylene, (poly) vinylidenedifluoride and polyacrylamide, as well as co-polymers and grafts thereof. A preferred solid support according to the invention is glass. The solid support can be any shape or size, and can exist as a separate entity or as an integral part of any apparatus (e.g. bead, cuvette, plate, vessel and the like) . It further is assumed that appropriate treatment of the solid support (e.g., glass) will be undertaken to provide adherence of polyacrylamide to the glass, e.g. with gamma- methacryl-oxypropyl-trimethoxysilane ("Bind Silane", Pharmacia) or other appropriate means in cases where the hydrogel is present on a solid support. In particular, covalent linkage of polyacrylamide hydrogel to the solid support can be done as de- scribed in EP 0 226 470. Preferably the solid support comprises or is made of a material selected from the group consisting of nylon, polystyrene, glass, latex, activated cellulose, glass, silica, modified silicon, ceramic, plastic, ZrO₂, Gold, polystyrene, polyethylene, polypropylene, (poly) tetrafluoroethylene, (poly) vinylidenedifluoride, polyacrylamide and combinations thereof. Furthermore the solid support is preferably selected from the group consisting of a bead, membrane, microwell, centrifuge tube, slide and sensor tip. Preferably the solid support can be treated with a coupling agent to attach, for instance, amine groups to its surface and the reactive site present on the solid support is attached to the solid support by the amine groups. Furthermore the surface of the solid support can be functional- ized to allow the immobilization of any chemical substances, especially biomolecules, in particular peptides, polypeptides, proteins, nucleic acids and the like, either directly or indirectly.

According to the present invention the solid support with the enzymes lipase/esterase are covered by a hydrogel matrix partially ("at least in part") or entirely. A partial covering of the solid support allows the provision of areas with different characteristics on said support. For instance, a biochip can have areas with a hydrogel matrix and areas with other surface properties, allowing the immobilization of other molecules on the same chip. The partial coverage of the solid support may result in the formation of two-dimensional or three-dimensional structures comprised of hydrogel (such as polyacrylamide) that is present on a solid support. Said "hydrogel structures" can be comprised of any variation of polymer (e.g. polyacrylamide or other polymer that is functionalized, activated, modified or otherwise combined with any appropriate moiety) such as is known in the art. Consequently, the solid support may be a "hydrogel array" having a combination of at least two of said structures . Preferably an array is comprised of hydrogel structures in addressable rows and columns. Also the thickness and the dimensions of the polymer hydrogel and/or hydrogel arrays produced according to the invention can vary dependent upon the particular needs of the user. However, the hydrogel structures may have a thickness of less than about 50μm. Furthermore, each of the hydrogel structures in an array may be from about 5 to about 1000 μm in size .

"Hydrogel matrix" refers to all hydrogel matrices regularly used in the art for similar purposes (immobilization of proteins) and known to a person skilled in the art. Hydrogels are generally defined as polymeric materials that swell in water and other fluids, absorbing the fluid within the polymer network without dissolving. Hydrophilic hydrogels have a large amount of water content at equilibrium and good biocompatibility . Furthermore hydrogels can be made sensitive to particular analytes, such as proteins, nucleic acids and the like. Hydrogels are suitably used in the scope of the present invention, because hydrogels have an adequate pore size and high water content to permit diffusion of molecules in and out of the matrix, are able to bind to the surface of a substrate, such as glass, have a sufficient transparency, in its fully polymerized state, to reduce optical interference with fluorescent tags, and have sufficient structural integrity, when fully polymerized, to withstand the forces encountered during use. Furthermore, the selected gel is preferably easy to produce and use.

Preferably, a "hydrogel" for use in the invention as a hydrogel is selected from the group consisting of polyamide, polyacrylamide, polyester, polycarbonate, polyvinylchloride, poly- methylacrylate, polystyrene and copolymers of polystyrene, polyvinyl alcohol, polyacrylic acid, polyethylene oxide, poly- urethane and polyvinylpyrrolidone. Preferably the polymer is polyacrylamide. Preparation of a hydrogel array (e.g. a polyac- rylamide hydrogel array) preferably comprises additional steps, optionally, developing the pattern in the array, and further optimally selectively removing the not cross-linked polymer in aqueous solution (e.g. water) to produce the hydrogel array, preferably the polyacrylamide hydrogel array. Pattern development desirably is accomplished by exposing the reactive prepoly- mer (e.g. a polyacrylamide reactive prepolymer) through a photomask. The hydrogel matrices preferably used in the present invention have a water content, at equilibrium, between 85% and 99%, preferably between 90% and 98%, in particular between 95% and 97%. Another important characteristic of hydrogels is the swelling capacity, because it reflects the maximum water content. Generally, the higher the water content of the polymerized gel, the faster the diffusion of molecules in and out of the gel. For instance, the hybridization reactions of DNA molecules in a hydrogel are more efficient the more water molecules are present in said hydrogel (Davis B. K. Proc. Natl. Acad. Sci. USA 21:3120, 1974).

The hydrogels used preferably to cover a solid support are gelatin hydrogels (e.g. US 6,815,078), dextran hydrogel (e.g. US 5,242,828), polyacrylamide (e.g. WO 03/014392, US 6, 664,061) , polyurethane (e.g. US 6,174,683).

In the WO 98/29736, an antibody microarray with antibody immobilized onto a N-hydroxysuccinimidyl ester modified glass substrate is disclosed.

In the US 5,981,734 and the WO 95/04594, a polyacrylamide based hydrogel substrate technology is described for the fabrication of DNA microarrays. More recently, the same hydrogel technology was further demonstrated as useful as a substrate for the immobilization of proteins in making protein microarrays (Aren- kov P et al. Anal. Biochem. (2000) 278:123-131).

Therefore, the hydrogel matrix preferably comprises dextran, gelatin, polyurethane, polyacrylamide, polyamide, polyacrylamide, polyester, polycarbonate, polyvinylchloride, polymethylac- rylate, polystyrene and copolymers of polystyrene, polyvinyl alcohol, polyacrylic acid, polyethylene oxide, polyvinylpyrrolidone or combinations thereof.

According to another preferred embodiment of the present invention the hydrogel matrix comprises a functional group selected from the group consisting of N-hydroxysuccinimide, strep- tavidin, biotin, antibodies, oligonucleotides, oligodesoxynuc- leotides and combinations thereof.

One of the key focuses in protein microarray is the ability to immobilize proteins in their native conformation on a solid support, while preserving their activities for functional studies. Several approaches have been developed to immobilize proteins, as well as other biomolecules, using either covalent attachments or non-covalent, affinity binding chemistries (Protein Microarray technology, Ed. Dev Kambhampati, Wiley-VCH, Weinheim, 2004) .

Therefore, in order to immobilize molecules to the hydrogel matrix, said matrix may comprise functional groups (modifications) . These functional groups are responsible for the binding of the enzymes to the hydrogel and consequently to the solid support. For instance, biotin may used to immobilize enzymes which comprise a streptavidin moiety resulting in a non-covalent binding. In contrast thereto, N-hydroxysuccinimide can be used to covalently bind an enzyme to the hydrogel . An antibody conjugated to the hydrogel in order to immobilize the at least one enzyme is preferably directed to the enzyme itself (preferably not to the active site of the enzyme) or to a enzyme moiety which is conjugated or recombinantly fused to said enzyme

The at least one enzyme is preferably conjugated to streptavidin, biotin, antibodies, oligonucleotides, oligodesoxynuc- leotides or combinations thereof.

The binding of the lipase/esterase to the hydrogel can be achieved by a chemical reaction or by conjugating a hydrogel specific ligand or binding partner to said enzymes. Specific binding partners or ligands may be preferably N-hydroxysuccinimide, streptavidin, biotin, antibodies, oligonucleotides and oli- godesoxynucleotides. The conjugation of said molecules to the enzymes may be achieved chemically or by recombinant technologies. Recombinant technologies are advantageously employed because such techniques allow to exactly position a ligand partner to the enzyme, thus, avoiding that the catalytical activity of the enzyme is negatively affected.

Another aspect of the present invention relates to an array comprising at least one compound inhibiting or potentially inhibiting at least one enzyme selected from the group consisting of lipase and esterase bound to a surface of a solid support, wherein the compound is an ester, preferably a p-nitrophenyl phosphonic ester.

According to the present invention not only lipases and esterases may be bound to a solid support, but also compounds binding or potentially binding to and, hence, inhibiting said enzymes. Said compounds are preferably conjugated (coupled) with biotin consequently allowing the binding to an appropriate solid support. Since fluorescently labeled p-nitrophenyl phosphonic acid esters are excellent activity recognition probes (ARPs) for the qualitative and quantitative determination of lipases and esterases, the compounds to be tested are preferably p-nitrophenyl phosphonic acid esters (see A 1273/2004) . It is therefore also possible to use these solid support hydrogel matrices for discriminating and analysing esterase and/or lipase activit- ies as well as for identification and/or characterization of esterase or lipase preparations, protein mixtures and extracts and complex samples comprising proteins, especially complex proteome samples .

Most lipases and esterases belong to the family of serine hydrolases that cleave carboxylic acid esters by a mechanism involving a well conserved catalytic triad and a nucleophilic serine residue. The nucleophilic serine is involved in the first reaction step leading to a tetrahedral intermediate state of the scissile fatty acid carbonyl group. This intermediate state is mimicked in a covalent and stable lipid-protein complex that is stoichiometrically formed by reaction of serine hydrolases with phosphonic acid esters inhibitors [8] . Fluorescently labeled p- nitrophenyl phosphonic acid esters are excellent activity recognition probes (ARPs) for the qualitative and quantitative determination of serine hydrolases [9] . These ARPs have already been widely used for identification and characterization of isolated enzymes and novel enzymes in complex proteomes [10-15] . Moreover, a rapid discrimination of lipase and esterase activities can be achieved when a library of structurally diverse ARP compounds is applied. The respective compounds are also excellent enzyme probes in chip technology.

Such compounds are also called suicide inhibitors. For instance, activity-based enzyme-inhibitor interaction on chip was already investigated using fluorescent suicide inhibitors for caspases, phosphatases, cysteine- and serine hydrolases [4-7] .

According to another preferred embodiment of the present invention said compounds are selected from the compounds according to 1 to 23 listed in figures 2A, 2B and 2C.

It turned out that especially these compounds bind to and react with lipases and esterases. An array comprising said compounds bound thereto may be used, for instance, in competition assays to determine the binding affinity of the enzymes to said compounds/ligands (see e.g. D. Leung et al., Nature Biotechnology 21 (6), 687-691 (2003); Eppinger et al., Angew. Chemie Intl. Ed. 43 (29), 3806-3810 (2004)).

Another aspect of the present invention relates to a method of determining qualitatively and quantitatively the binding of a compound to an enzyme selected from the group consisting of lipase and esterase comprising the steps : a) providing

I) a protein array comprising said enzyme or

II) an array comprising at least one compound inhibiting or potentially inhibiting said enzyme bound to a solid support, b) contacting array I) with at least one compound inhibiting or potentially inhibiting said enzyme or contacting array II) with said enzyme forming an inhibitor/enzyme complex, and c) detecting binding of the at least one compound to said enzyme.

In order to determine the binding of a molecule to a lipase/esterase a protein array comprising a lipase or an esterase or an array comprising potential binding partners (like inhibitors) is contacted with a sample comprising a potential binding partner or a sample comprising catalytically active lipase or esterase, respectively. If both binding partners exhibit an affinity to each other, a molecule (inhibitor) /enzyme complex is formed. Said complex can be detected by antibodies directed to the lipase or esterase or to the compound employed in the method according to the present invention. To facilitate such a binding, the enzymes or the compounds may comprise a functional group. Furthermore it is also possible to detect said complex when in the course of the interaction the compound releases (due to a cleavage reaction) a compound moiety which is detectable, for instance, with a fluorometer or a photometer.

The protein array I) and array II) are preferably arrays according to the present invention.

According to a preferred embodiment the at least one compound is an ester, preferably a p-nitrophenyl phosphonic ester, and selected preferably from the compounds according to 1 to 23 listed in figures 2A, 2B and 2C.

The at least one compound is preferably biotinylated.

According to a preferred embodiment of the present invention the binding of the at least one compound to the lipase and/or esterase is detected by contacting the array with a first antibody directed towards the lipase, the esterase or the at least one compound and optionally with a second antibody directed to said first antibody, wherein the first or second antibody is labeled with an enzyme, a fluorophore, a biotin, an enzyme substrate and any other affinity tag. The enzyme, which is coupled to the first or second antibody, is preferably selected from the group consisting of horseradish peroxidase, alkaline phosphatase and luciferase.

The fluorophore is preferably selected from the group cyanines, merocyanines, boradiazaindacene, rhodamins, anthracenes, nitrobenzoxadiazoles and combinations thereof.

Another aspect of the present invention relates to a kit for determining binding of a compound to an enzyme selected from the group consisting of lipase and esterase comprising:

- an array covered by a hydrogel as defined in the present invention,

- a lipase and/or an esterase and

- optionally means for detecting the binding of the compound to said enzyme.

The kit according to the present invention can be employed, for instance, in a method as described above.

The present invention is further illustrated by the following figures and examples, without being restricted thereto.

Figure 1: Microarrays for activity-based lipase and esterase profiling. A: Enzymes immobilized on chip can be screened with a given fluorescent activity recognition probe (ARP) in solution; B: ARPs immobilized on chip are probed with a given enzyme in solution.

Figure 2 (2A to 2C) : Fluorescent activity recognition probes (ARP) (according to WO 2006/010403 and WO 2004/083220) .

Figure 3: Effect of inhibitor concentration on activity- based fluorescence labelling of MME on chip. 400 μg/ml MME (concentration used for spotting) was spotted on chip (corresponding to ~ 36 pg protein per spot) and reacted with 100 μl (1-1000 nM) of ARP for 30 min. ARPs: NBD-HE-HP 1 (■) ; NBD-Cl2-Biotin 2 (•) (for chemical structures see figures 2A to 2C) .

Figure 4: Effect of protein concentration on reaction of immobilized MME with ARPs. 5-500 μg/ml MME (concentration used for spotting, corresponding to approximately 4.5 to 450 pg protein /spot) was spotted onto the chip and reacted with 100 μl of 100 nM activity-based probe for 30 min. Inhibitors: A: NBD-HE-HP 1 (■) ; B: NBD-C12Biotin 2 (•).

Figure 5: Time dependent labelling of MME on chip by fluorescent ARPs. Direct detection of ARP fluorescence on chip: NBD- HE-HP 1 (■, panel A) NBD-C12-Biotin 2 (•, panel B) . Indirect de- tection of NBD-C12-Biotin 2 by Avidin-ALEXA 555 conjugate is shown in panel C (o) . Inhibitor concentration in the incubation solutions was 5OnM.

Figure 6: Deactivation of immobilized MME. MME arrays were consecutively incubated with the indicated reagents as follows: 1. Incubation with 7 M urea/2 M thiourea; 2. Probing the enzymes with the inhibitors; 3. Detection of biotinylated inhibitor 2 using avidin-ALEXA 555 conjugate (dilution: 1:500); 4. Incubation with 7 M urea/2 M thiourea. ARP concentrations in the incubation solutions were: 1000 nM (a) and 100 nM (b) For chemical structures of NBD-HE-HP 1 and NBD-C12-Biotin 2 see figure 2. Fluorescence detection: NBD: (lex: 480nm; lem: 530nm) , avidin- ALEXA 555 conjugate (lex: 532nm; lem: 595nm) . Incubation volume: lOOμl of ARP solutions. Incubation time for denaturation: 15 min.

Figure 7: Reaction of ALEXA 647-labeled Mucor miehei esterase with different concentrations of immobilized ARP 2 on chip. Streptavidin-coated slides were incubated with ARP 2 at concentrations ranging from 50 μM (-1.7 on x-axis) down to 10OfM (7 on x-axis) . Finally slides were incubated with lOOμl of the labeled enzyme (14μg/ml), followed by washing and fluorescence scanning (A, lex: 480nm; lem: 530nm, fluorescence of ARP and B, lex: 635nm; lem: 685nm, fluorescence of labeled enzyme).

Figure 8; Binding of ALEXA 647 labeled MME and MGL to ARPs 2 and 20-23 (Figure 2) immobilized on streptavidin-coated slides. MME: Slides were incubated with 100 μl of the labeled MME (14 μg/ml) at 37 °C for 30 min, followed by washing and scanning of fluorescence intensities (false colored red: lex: 635nm; lem: 685nm) . For measurement of the blank fluorescence, ARP binding to streptavidin was omitted prior to incubation with MME. MGL: Slides were incubated for 60 min with 100 μl of the crude cytoplasmic lysate (1.11 mg/ml, total protein) of COS-7 cells transiently expressing murine MGL. Bound MGL was identified immunologically coupled to enzymatic hydrolysis of ELF-97 phosphate, yielding a fluorescent precipitate which was scanned at lex: 360nm; lem: 530nm (false colored green. Blank fluorescence intensity was obtained after incubation of cytoplasmic extracts (1.14 mg/ml, total protein) from COS-7 cells transfected with Lac Z in the same vector (pcDNA4/HisMax-Invitrogen) under the above described conditions. Figure 9: Array layout: Arrays consist of 6 subarrays representing the 6 enzymes under investigation each spotted 6-fold in 9 different concentrations.

EXAMPLES :

Example 1: Immobilization of lipases/esterases on a solid support

Proteins were spotted in print buffer (30OmM phosphate buffer pH = 8.50, 0.005 % CHAPS and 1 % Trehalose) onto Schott Nex- terion slide H microarray slides (Schott Nexterion, USA) at 25⁰C and 45 % relative humidity using SMP 3B stealth pins (Telechem, USA) in combination with a QArraymini robot (Genetix, UK) and Qsoft microarray software. During spotting, the source mi- crotiter plate was kept at 6°C to reduce water evaporation and preserve enzyme activity. Each enzyme was spotted 6-fold at 500, 400, 300, 200, 100, 50, 25, 10 and 5 μg/ml (corresponding to approximately 450, 360, 270, 180, 90, 45, 22.5, 9 and 4.5 pg/spot) to generate an 18 x 18 array with a spot to spot pitch of 300 μm. For preparation of streptavidin microarrays 500 μg/ml streptavidin in print buffer was spotted onto Nexterion SlideH slides to generate 18 x 18 arrays. Twelve of these arrays were printed onto one slide in a 6 x 2 pattern which was compatible with 16-pad incubation chambers of the FAST FRAME system (Schleicher & Schuell, Germany) . To ensure proper protein binding, printed slides were incubated at 75 % relative humidity and 25⁰C for 1 hour. Slides were used immediately or stored at 4⁰C for a maximum of two days .

After spotting and incubation at 75 % humidity, the remaining active surface of the slides was blocked with 5OmM ethano- lamine in 5OmM borate buffer pH = 8.0 at 25⁰C for 1 hour prior to incubation with the ARPs in solution. The blocking step was performed immediately before incubation of the slides with ARPs in order to protect the proteins from dehydration.

Example 2: Enzyme inhibitor studies using a Mucor miehei esterase array

Enzymes in active form were immobilized on N-hydroxysuccin- iminic acid (NHS) activated hydrogel slides (Nexterion Slide H, Schott Nexterion, USA) . The ARPs used for probing the enzymes on chip were the same that have been used for lipase and esterase detection in homogeneous systems.

In a first approach identical concentrations of commercially available Mucor miehei esterase (MME, Fluka, Germany) were spotted and probed with different concentrations of inhibitors 1 and 2 (for structures see figures 2A to 2C) ranging from 1-100OnM (see figure 3) .

Basically two systems, the protein microarray and the small molecule microarray, can be designed to study activity-based interaction of enzymes and inhibitors. The activity of a single inhibitor towards a group of different enzymes is determined using a microarray containing different immobilized enzymes that are probed with a single ARP in solution (figure 1 panel A) . The preference of a certain enzyme for different ARPs can be studied by probing the immobilized ARPs with the enzyme in solution (figure 1, panel B) .

Immobilized MME was inhibited in a concentration dependent manner by both ARPs. Fluorescence intensities (relative fluorescence units, RFU) leveled off at 25nM (ARP 1) and 50OnM (ARP 2). The detection limit was 1 nM for both inhibitors. Conversely, different enzyme concentrations of MME (5-500 μg/ml) were spotted and incubated with 100 nM ARP in solution. A concentration- dependent interaction of MME with the individual probes was observed for ARP 1. Binding of Inhibitor 2 only reached saturation at lOOμg/ml. The minimum amount of MME detectable by ARPs on mi- croarrays was 50 μg/ml in the spotting solution. This amount is equivalent to approximately 45 pg of enzyme per spot assuming a spot delivery volume of 0.9 nl per stamp and complete sample binding on the surface of the slide (figure 4) [17] . A fluorescence intensity 3D-plot of spots showed that the heights of the spots did not increase linearly. Therefore, the plotted RFUs were calculated from the background-corrected spot volume, which did not correlate with spot height.

The time course of the reaction of ARPs with MME (figure 5) revealed that the different time-dependent reactivities of the inhibitors with the protein on chip could be well determined. Reaction of MME with ARP 1 reached completeness after 60 min whereas reaction with inhibitor 2 levelled off after two hours. Inhibitor 2 has a biotin-moiety in addition to a fluorescent tag (figure 2) . Thus, it can either be detected on the basis of its fluorescence signal, or alternatively, after addition of fluor- escently labeled avidin. However, indirect detection of inhibitor 2 by an avidin-ALEXA 555 conjugate only poorly reflected the reaction progress as compared to direct fluorescence detection (panel C) . The optimal time window for kinetic discrimination of enzyme reactivity on chip by ARP 1 and 2 was between 30 and 60 min. In this range enzymes labelling by the inhibitor, and changes of signals linearly depended on time. Consequently, all incubation times for chip screening with ARPs were set to 30 min throughout these examples unless otherwise indicated.

In order to confirm that the ARPs solely detect active enzymes, MME on chip was denatured prior to incubation with ARPs using a mixture of 7M urea / 2M thiourea at 37⁰C for 15 min (figure 6) . After denaturation, binding of the ARPs to the spotted proteins was no longer observed, demonstrating that the ARPs solely bind active enzymes. Indirect detection of inhibitor 2 binding to MME on chip by the fluorescent avidin conjugate was less sensitive as compared to direct measurement of the NBD- fluorescence of this enzyme probe. Consecutive incubation with the denaturation agent urea / thiourea diminished indirect detection by the avidin-ALEXA 555 conjugate and thus reduced the binding of biotin to avidin. Incubation of MME on chip with 10OnM of ARP 2 was not sufficient for indirect detection of bound probe by the biotin avidin-ALEXA 555 conjugate system whereas fluorescence of the NBD-tag was well distinguishable from background. The same observation was made when fluorescent ARPs were compared to biotinylated derivatives in proteome profiling experiments [18] . Treatment of MME on chip with sample buffer for SDS-polyacrylamide gel electrophoresis at 37⁰C for 15 min did not abolish activity-based enzyme recognition. Instead, heating the slides to 95⁰C for several min was mandatory for enzyme deactivation.

Example 3: Enzyme inhibitor studies using an array of 6 different lipolytic enzymes

For parallel enzyme activity profiling on chip, 6 different lipolytic enzymes (table 1, for preparation of the enzymes see Example 6) at 9 different concentrations between 5-500μg/ml (spotted volume ~ 0.9nl, corresponding to ~ 4.5 - 450 pg enzyme per spot) were spotted. One microarray slide contained 6 X 2 identical arrays consisting of 18 X 18 features in which every enzyme concentration was spotted six-fold. 16-pad incubation chambers in combination with the FAST FRAME system (Schleicher & Schuell) were used. Thus, up to 12 independent experiments could be performed with a single microarray slide (see figure 9) .

The fluorescently labeled ARPs that were used to detect active lipases and esterases immobilized on the microarrays are depicted in figures 2A to 2C.

The immobilized enzymes (4.5 - 450 pg / spot) on chip were probed with 100 μl of a 10OnM solution of the ARPs in 1OmM Tris- HCl pH = 7.40, 0.25M sucrose and ImM Triton X-100 at 37⁰C for 30 min. After incubation, slides were washed and fluorescence was scanned at 530nm. For comparison of labelling by the different ARPs, the fluorescence intensity readouts of the array scanner for each individual ARP and enzyme were expressed as fractions of the fluorescence intensity of the ARP showing the highest intensity (reactivity) when bound to the same enzyme.

Table 1: Relative fluorescence intensities of ARP-labeled enzymes. Values are given as the percentage of the fluorescence readout of an enzyme-bound ARP divided by the readout of the best bound ARP by the same enzyme. Inhibitor concentration in the incubation mixture: 10OnM. bCE: Bovine cholesterol esterase; CAL B: Candida antarctica lipase B; CLE: Candida lipolytica esterase; ppL: porcine pancreatic lipase; MME: Mucor miehei esterase. A: enzyme reactivity on chip; B: enzyme reactivity in solution; n.d.: not determined

Example 4: Preparation of activity recognition probe (ARP) solutions and a ARP arrays

The ARPs (according to A1273/2004 and A440/2003) were stored solvent-free at < -20 °C. Aliquots of ARPs dissolved in chloroform: methanol 2:1 (v/v) were transferred to an Eppendorf tube followed by addition of 15μl of 1OmM Triton X-100 in chloroform. The organic solvent was removed under a gentle stream of argon and the solvent-free residue was dissolved in 150μl of assay buffer (10mm Tris/HCl pH = 7.40, 0.25 M sucrose) by vortexing for 5 min. For the streptavidin-arrays, the end concentration of the incubation solution was adjusted to 10 mM Triton X-100 and 5 μM biotinylated ARPs. Subsequently, the prepared ARP solutions were transferred to a 96 well microtiter plate, incubated on an Eppendorf shaker at 37⁰C and 550 rpm for 5 min. lOOμl of the individual samples were layered onto the microarray using an 8 channel transfer pipette. The ARPs were incubated with the array at 37°C for 30 min unless otherwise indicated. After extensive washing with TBST (20 mM Tris pH = 7.40, 15OmM NaCl, 1% Tween- 20) , and brief rinsing with deionized water, NBD fluorescence of the bound inhibitors was scanned (lex: 480nm; lem: 530nm; exposure time: 0.1 s) using an ArrayWorxe Auto microarray scanner (Applied Precision, Issaquah, WA) .

The biotin moiety of ARP 2 was detected after incubation with Avidin-ALEXA conjugate (1:500 (v/v) in TBST (1% Tween-20)) at room temperature for 30 min. The label fluorescence was scanned after extensive washing of the chips with TBST (1% Tween-20) , followed by brief rinsing with deionized water (lex: 540nm; leirv: 595nrα; exposure time: 0.25 s) .

Example 5: Enzyme binding assay to small molecule (binding partners, inhibitors) microarrays

The small molecule microarrays (ARPs immobilized on strep- tavidin slides, see e.g. Example 7) were incubated with MME or MGL at 37⁰C for 30 and 60 min, respectively. After incubation, the microarrays were extensively washed with TBST (0.1 % Tween- 20) . Slides containing bound MME-ALEXA 647 conjugate were directly imaged (lex: 635nm; lem: 685nm; exposure time: 3 s) . Slides that had been probed with COS-7 cell lysates were incubated with anti-6 x HIS antibody (1:150 (v/v) ; Amersham Bios- ciences, raised in mouse) in TBST (0.1 % Tween-20) at room temperature for 30 min. After extensive washing with TBST (0.1 % Tween-20), anti-mouse antibody-alkaline phosphatase conjugate (1:150 (v/v); Sigma, raised in goat) in TBST (0.1 % Tween-20) was incubated for 30 min. Slides were consecutively washed with TBST (0.1 % Tween-20) and alkaline phosphatase reaction buffer (APRB; 1OmM Tris pH = 9.50, 1OmM NaCl, 0.5mM MgCl2, 0.05 % Triton X-100) . Finally, the slides were incubated with 25OnM ELF-97 phosphate in APRB at room temperature for 15 min. The reaction on the chip was stopped by extensive washing with TBST (0.05 % Triton X-100) followed by fluorescence scanning of the generated precipitate on the chip surface (lex: 360nm; lem: 530nm; exposure time: 0.02 s) .

All experiments were performed in duplicate and microarray data were analyzed using Genepix pro 5.1 software (Axxon) .

Microarrays of lipase and esterase inhibitors were developed as a tool for rapid screening of enzyme preference for different substrate analogous ARPs. In a first attempt, NHS-activated precursors of the ARPs listed in figures 2A, 2B and 2C were immobilized directly to amino modified hydrogel slides. In another experiment, microarrays of biotinylated ARPs on streptavidin- coated hydrogel slides were generated (figure 1, panel B) . Instead of a fluorophore, these compounds contained a biotin tag for the reaction with a streptavidin coated chip surface. The ARPs in this experiment bind exclusively to streptavidin via their biotin moiety. No covalent binding of the phosphonic acid ester was detected in control experiments. For detection of enzyme binding to such chips, the proteins had to be labeled prior to incubation under conditions that did not affect proper enzyme function. In a typical experiment 500μg/ml (450 pg / spot) streptavidin was spotted in subarrays containing 18 x 18 features. After blocking of the remaining active NHS-esters on the array surface, slides were incubated with the biotinylated ARPs. Unbound ARPs were washed off, and the ARP microarray was incubated with the fluorescently tagged lipase or esterase. For the experiments chips were used on which different concentrations of ARP 2 were spotted prior to incubation with MME-ALEXA 647 conjugate. The amount of ARP 2 immobilized on streptavidin-coated slides and the activity-based binding of MME to these microarrays was simultaneously monitored by measurement of lipid (NBD) and protein (ALEXA 647) fluorescence, respectively (figure 7). The fluorescent signal remained constant over a concentration range of 4.5 orders of magnitude. At very low amounts of ARP 2 (10OfM) binding of MME was reduced. The relative amounts of bound enzyme was determined from the ratio of fluorescence intensity at 685nm (MME-ALEXA 647) divided by the intensity at 530nm (NBD-ARP) . The ratio increased linearly from 25nM to IpM, and decreased again at 10OfM ARP. Under these conditions, ARP fluorescence intensity in the 685nm channel was already much lower as compared to the background intensity in the 530nm channel.

The activity-based inhibition of labeled MME by different biotinylated ARPs (figures 2A to 2C) immobilized on streptavidin slides is shown in figure 8. The biotinylated ARP 22 was the best probe for MME. Except for the label, the fluorescent derivative 11 (NBD-tag) is structurally identical to ARP 22 (biotin- tag, figure 2B) . It was the best MME inhibitor under investigation either on chip or in solution. The triacylglycerol-analog- ous ARPs 20 and 21 in immobilized form were also recognized by MME (figure 8) . In contrast, the fluorescently tagged ARPs 3 and 4 did not react with immobilized MME (table 1) . To demonstrate the capacity of the ARP microarrays of detecting active proteins, binding of N-terminal 6 X His-tagged murine monoacylgly- cerol lipase (MGL, overexpressed in COS-7 cells) to the small molecules on chip was studied using crude cytoplasmic extracts of MGL -transfected COS-7 cells. MGL was previously identified as a target of several activity recognition probes in many pro- teomes [19]. The crude cytoplasmic extracts (1.11 mg/ml total protein) were incubated with small molecule microarrays containing ARP 2, 20 - 23 for 1 hour. The His-tag of the bound enzyme was immunologically detected by antibody-alkaline phosphatase conjugates followed by hydrolysis of ELF-97 phosphate (Molecular Probes, Leiden) and measurement of precipitate fluorescence on chip [20] .

Example 6: Enzyme preparations to be used in inhibitor studies using lipase/esterase arrays

Bovine cholesterol esterase (bCE, Sigma, C-3766, Lot 119H7405); Mucor miehei esterase (MME, Fluka, 46059 Lot: 10249/2); Candida lipolytica esterase (CLE, Fluka, 46056 Lot: 48839/1); Candida antarctica lipase B (CAL B, Novo, SP525, Lot: PPW 5328) ; Chromobacterium viscosum lipase (CVL, Biocatalysts, Lot: 3169/2484); porcine pancreatic lipase (ppL, Sigma, L-0382, Lot:031K7670) ; Streptavidin (Sigma, S-4762, Lot: 118H86605) ; Enzymes were dissolved in distilled water followed by tranfer to print buffer (30OmM phosphate buffer pH = 8.50, 0.005 % CHAPS and 1 % Trehalose) using PD-IO desalting columns (Amersham Bios- ciences, Austria) . Protein concentration was determined using the Bio-Rad protein assay reagent based upon the method of Bradford. The final protein concentration was adjusted to 2 mg/ml using print buffer.

Lyophilized streptavidin was directly dissolved in print buffer at 1 mg/ml. Enzyme stock solutions were stored at -20⁰C. Transient expression of mouse monoacylglycerol lipase (MGL) was performed as described.

MME and avidin were labeled with ALEXA 647-NHS and ALEXA 555-NHS (Molecular Probes, The Netherlands), respectively. Protein labelling and purification of the conjugate was performed as recommended by Molecular Probes .

MGL was transiently expressed in COS-7 cells. Transfection of COS-7 cells was performed with Metafectene™ (Biontex, Germany) according to the manufacturer's instruction. The apparent molecular weight of His-tagged MGL was 33 kDa, as confirmed by Western blotting using an anti-His monoclonal antibody (6 x His, Amersham Biosciences) at a dilution of 1:7000. Transfected COS-7 cells were washed twice with PBS, scraped into lysis buffer (0.25 M sucrose, 1 mM EDTA, 1 mM dithioerythritol, 20 μg/ml leu- peptin, 2 μg/ml antipain, 1 μg/ml pepstatin) and lysed by brief sonication (Virsonic 475, 4mm sonication tip) at maximum power on ice for 10 s. Nuclei and unbroken material were removed by centrifugation at 1.000 g at 4⁰C for 15 min to obtain cytoplasmic extracts.

Example 7: Binding of biotinylated ARPs to streptavidin

In biotinylated ARPs, the biotin residue anchors the molecule probe to the streptavidin on chip, whereas the phosphonic acid ester moiety specifically reacts with the nucleophilic serine in serine hydrolase enzymes. To exclude the possibility that biotinylated ARPs bind to streptavidin via the phosphonate moiety (covalent binding) the following control experiment was performed. Aliquots of samples containing 20 μg (0.29nmol) streptavidin (binds 1.15 nmol biotin) were incubated with 2 nmol ARP 2, 20-24, respectively, in 20 μl 1OmM Tris-HCl pH = 7.40, 1OmM Triton X-100 at 25⁰C for 1 hour. After incubation, 5μl of a 5 fold concentrated ID SDS-PAGE sample buffer was added, and samples were incubated at 95⁰C for 5 min.

Proteins were separated by ID SDS-PAGE, electro-blotted on nitrocellulose membranes (Schleicher & Schuell, Germany) and blocked with 5 % non-fat milk powder in TBST (1% Tween-20) . Biotinylated proteins were detected using an avidin ALEXA 555 conjugate (dilution of 1:2000 (v/v) in TBST (1% Tween-20)).

Example 8: Discussion of Examples 1 to 7

The immobilization and detection of enzymes in active form is a key step towards functional protein microarrays. Very much progress has already been made in chemical surface modification, especially of glass substrates, that is the basis for preparation of enzyme microarrays. It surprisingly turned out that hy- drogel-coated slides gave the best results when used for immobilization of enzymes, especially lipases/esterases, in active form. These systems were superior to enzyme chips based on epoxy, amino, or dendrimer surfaces [7;16;21]. Chen et al. reported that hydrolytic enzymes can be maintained in active form on epoxide-activated slides. Their approach was successful when the enzymes were proteases but it did not work with lipases [6]. According to the experimental data provided herein, hydrogel- coated slides now open up the possibility for functional screening of active lipases and esterases on chip. For enzyme recognition phosphonic acid inhibitors that have already been successfully used for lipase and esterase screening in solution were used. The inhibition of immobilized enzymes by these activity recognition probes is linearly dependent on ARP concentration (figure 3), concentration of immobilized enzyme (figure 4) and time (figure 5) . Labelling of the same amount of immobilized enzyme (MME) with ARP 1 and 2 in solution, reached saturation (same fluorescence intensity) at different inhibitor concentrations. Saturation with ARP 1 was observed at about 5OnM, whereas ARP 2 binding leveled off at 50OnM. The saturation of lipid protein binding at higher ARP concentrations in conjunction with the resistance of the bound label to solvent extraction may be taken as proof that the reaction between the activity recognition probes and the immobilized proteins was covalent and specific, and had nothing to do with plain surface adsorption. According to the initial slopes of the binding curves as depicted in figure 2, ARP 1 has a 25-fold higher affinity to immobilized MME than ARP 2. Figure 4 shows the relationship between the concentration of immobilized MME and the extent of inhibitor binding (10OnM ARP 1 and 2 in solution) showed a linear dependency in the concentration range from 50 to 500 μg/ml (45-450 pg enzyme per spot) for ARP 1 [17] . A twofold increase in fluorescence intensity due to binding of the labeled lipid was associated with the same increase in immobilized enzyme concentration on chip indicating that the entire amount of enzyme spotted was accessible for the reaction with the inhibitor. In contrast, enzyme labelling by ARP 2 was a linear function only below lOOμg/ml (90pg enzyme per spot) . Apparently, only a limited fraction of the spotted enzyme was accessible for this more hydrophobic inhibitor. However, enzyme labelling linearly increased when increasing concentrations of ARP 2 in solution were applied. This effect was especially observed when hydrophobic ARPs (2-5; 12 and 13) were used as probes. The time-dependent reaction of ARPs 1 and 2 with immobilized MME again reflects the higher affinity of ARP 1 as compared to ARP 2. The initial reaction rate was 3 fold faster and reached saturation at 60 min incubation time with ARP 1, whereas saturation with ARP 2 required 2 hours. The observed difference of MME reactivity towards ARP 1 and 2 can be explained by the fact that ARP 1 is less hydrophobic and thus, is preferred as a substrate by the microbial esterase.

For activity-based enzyme screening, it is important that the inhibition reaction has not reached completeness at the time the slide is prepared for fluorescence readout (removal of the chip from the incubation solution) . According to previous experiments it was shown that ARP 1 is not specific and shows high affinity towards many lipolytic enzymes. Therefore, this inhibitor and ARP 2, in combination with MME, were chosen for optimization of reaction conditions that should apply to the general analysis of most of the known lipolytic enzymes. In these experiments, enzymes of microbial and animal origin were screened for their capacity of reacting with ARP 1 and 2, and all of them were active under the optimized protocol. Thus, this procedure might be also used for screening of lipolytic activities in unknown purified protein samples.

Table 1 shows the relative reactivities of different inhibitors in solution for different lipases and esterases on hydro- gel chips. In addition, comparable data are provided for inhibitor reactivities towards bovine cholesterol esterase (bCE) , porcine pancreatic lipase (ppL) and Mucor miehei esterase (MME) in a solution phase assay. In general, the open chain ARPs 1-2 and 6-10 show considerable reactivity for the enzymes on chip, although there is significant discrimination between the individual enzymes. The bulky, hydrophobic ARPs 3-5 show extremely low or now affinity at all. ARPs 14-19 which contain heterocyclic rather than lipidic moieties in close vicinity to the reactive phosphonate centers are not very reactive towards enzymes on chip either. The relative reactivities of the immobilized enzymes bCE, ppL and MME are not fully comparable to the values obtained for the enzymes in solution. Binding of the very hydrophobic inhibitors 3-5 to the immobilized enzymes was modest in both systems. In addition to poor enzyme labelling, the background signal due to unspecific surface staining was dominant when inhibitors 3-5 were solubilized in ImM Triton X-IOO. A 10- fold increase in Triton X-IOO concentration did not reduce the background signal. Therefore, the apparent binding of ARPs to the enzymes on chip is both a function of enzyme-ARP affinity and ARP solubilization in the aqueous incubation buffer, which in turn depends on ARP hydrophobicity. MME affinities for ARPs were very similar on chip and in solution except for the very hydrophobic inhibitors 3-5. Inhibitor affinities of bCE were only comparable for ARPs 6-11 representing the rather polar com- pounds. In summary, the observed enzyme activities on chip do not in all cases match their functional properties in solution. Nevertheless, the present data suggest that microarrays of immobilized lipases and esterases are well defined systems that can be used for at least qualitative or semi quantitative, activity-based, enzyme screening in a high-throughput format.

Whereas enzyme microarrays are useful tools for screening protein activity towards a given (substrate analogous) inhibitor, ARP microarrays could be applied to parallel determination of the preference of a given enzyme for the individual small molecules on chip. In addition, ARP arrays might help circumvent a general problem associated with protein immobilization on surfaces. The immobilization of enzymes on solid supports is a critical step, and often leads to a decrease in activity. Thus, the small molecule ligands were immobilized by spotting biot- inylated activity recognition probes onto streptavidin-coated glass slides. Specifically, binding of fluorescently tagged MME to the biotinylated ARP 2 on chip was studied. The extent of lipid-protein binding was the same within an ARP concentration range of 4.5 orders of magnitude. Covalent protein-ARP binding is stoichiometric (1:1 mol/mol) and once an enzyme molecule is bound, many ARPs present in a single spot are not accessible due to sterical reasons. As a consequence, the ARP density in a spot that is required for capturing a maximum amount of enzyme is rather low. Small molecule microarrays were generated by spotting 5 different biotinylated inhibitors onto streptavidin coated slides, and were successfully used for functional screening of ALEXA647-tagged MME in solution. MME discriminated between the individual ARPs, showing the highest affinity for inhibitor 22. In addition, the concentration of MME in solution correlated very well with the fluorescence intensities on chip after incubation. Reduction of enzyme concentration in the incubation solution phase by a factor of 2 (from 14 to 7 μg/ml) led to a decrease in fluorescence intensity readout by the same factor. labeled enzymes are needed for detection of protein binding on small ligand arrays, when antibodies are not available for such enzymes. Proteins can be properly labeled with a minimum amount of fluorophore, maintaining the functional properties of the enzyme. Today many enzymes are expressed as recombinant pro- teins in suitable host systems. In order to simplify enzyme purification as well as identification of the overexpressed protein in the host, recombinant enzymes are often expressed as affinity-tagged fusion proteins. These tags can also be used for identification of enzymes if bound to a small molecule microar- ray, e.g. due to covalent binding by the spotted activity recognition probes. Here, the chip binding of crude cytoplasmic extracts from COS-7 cells containing transiently expressed MGL tagged with 6 x His is described. When an ALEXA-647 labeled anti-mouse antibody was used for detection of the primary anti-6 X His antibody, sensitivity was too low to measure binding of the associated MGL. However, binding was successfully visualized upon detection of the His-tag by a signal amplification procedure. For this purpose, an anti-mouse-alkaline phosphatase conjugate was used as secondary antibody followed by incubation with ELF 97-phosphate . The latter compound is an alkaline phosphatase substrate which upon enzyme-mediated hydrolysis forms a fluorescent precipitate.

This system is being widely used for immunochemistry, Western blot analysis, etc. In the prsent invention it is described for the first time as a detection tool in chip technology. In summary, a powerful microarray-based system for the rapid screening of ARP binding to immobilized lipolytic enzymes is developed and, vice versa, of protein binding to different ARPs immobilized on chip. This system might also become useful for other biologically relevant effects including the action of competitive inhibitors, substrates and protein-protein interactions. Hydrogel slides turned out to be the most suitable surface for enzyme chip technology in our hands. Enzymes as well as ARPs could be deposited in functional form on these systems. Functional ARP chips were alternatively produced by spotting bi- otin-tagged derivatives on streptavidin-spotted slides. Enzyme- ligand binding was detected either by direct measurement of label fluorescence or, if enzymes were captured from solution, by fluorescent signal amplification. The above techniques were applied to functional analysis of lipases and esterases of different origin (microbial and animal) and substrate preference (tri- acylglycerol-, monoglycerol-, cholesterol ester and carboxylic acid ester hydrolases) . Continuing efforts are now being made to optimize routine applications of biochips for enzyme analysis in biotechnology and biomedicine.

References

[1.] M. F. Templin, et al . Trends in Biotechnology 2002, 20:160-

166.

[2.] G. MacBeath, Nature Genetics 2002, 32:526-532.

[3.] D. S. Wilson, S. Nock, Current Opinion in Chemical Biology

2002, 6:81-85.

[4.] M. Uttamchandani, et al . Bioorganic & Medicinal Chemistry Letters 2003, 13:2997-3000.

[5.] Q. Zhu, et al. Organic Letters 2003, 5:1257-1260.

[6.] G. Y. J. Chen, et al. ChemBioChem 2003, 4:336-339.

[7.] J. Eppinger, et al . Angewandte Chemie (International Ed. in English) 2004, 43:3806-3810.

[8.] U. Derewenda, et al . Biochemistry 1992, 31:1532-1541.

[9.] O. V. Oskolkova, et al. Chem. Phys .Lipids 2003, 125:103-114.

[10.] D. A. Campbell, et al. Current Opinion in Chemical Biology

2003, 7:296-303.

[11.] R. Zimmermann, et al. Science 2004, 306:1383-1386.

[12.] G. C. Adam, et al. MoI Cell Proteomics 2002, 1:781-790.

[13.] Y. Liu, et al . PNAS 1999, 96:14694-14699.

[14.] D. Leung, et al . Nature Biotechnology 2003, 21:687-691.

[15.] D. Kidd, et al. Biochemistry 2003, 40:4005-4015.

[16.] P. Angenendt, et al . J Chromatogr A 2003, 1009:97-104.

[17.] delivery volume of 0.9 nL for SMP3B split pins was taken from www.arrayit.com. 20-5-0005.

[18.] M. P. Patricelli, et al . Proteomics 2001, 1:1067-1071.

[19.] N. Jessani, et al . PNAS 2005, 101:13756-13761.

[20.] R. P. Haugland, in Handbook of Fluorescent Probes and Research ProductsEd. : J. Gregory) , Molecular Probes, Eugene, OR

2005, p. pp. 159-166.

[21.] S. A. Sieber, et al . J.Am. Chem. Soc . 2004, 126:15640-15641.

Claims

Claims :

1. A protein array comprising at least one enzyme selected from the group consisting of lipase and esterase on a solid support, characterized in that the enzymes on said solid support are covered at least in part with a hydrogel matrix.

2. Array according to claim 1, characterized in that the hydro- gel matrix comprises dextran, gelatin, polyurethane, polyacryl- amide, polyamide, polyacrylamide, polyester, polycarbonate, polyvinylchloride, polymethylacrylate, polystyrene and copolymers of polystyrene, polyvinyl alcohol, polyacrylic acid, polyethylene oxide, polyvinylpyrrolidone or combinations thereof.

3. Array according to claim 1 or 2, characterized in that the solid support comprises a material selected from the group consisting of nylon, polystyrene, glass, latex, activated cellulose, glass, silica, modified silicon, ceramic, plastic, ZrO₂, Gold, polystyrene, polyethylene, polypropylene, (poly) tetra- fluoroethylene, (poly) vinylidenedifluoride, polyacrylamide and combinations thereof.

4. Array according to any one of claims 1 to 3, characterized in that the hydrogel matrix comprises a functional group selected from the group consisting of N-hydroxysuccinimide, streptavidin, biotin, antibodies, oligonucleotides, oligodesoxynucleotides and combinations thereof.

5. Array according to any one of claims 1 to 4, characterized in that the at least one enzyme is conjugated to streptavidin, biotin, antibodies, oligonucleotides, oligodesoxynucleotides or combinations thereof.

6. An array comprising at least one compound inhibiting or potentially inhibiting at least one enzyme bound to a surface of a solid support, characterized in that the compound is conjugated to biotin or is an ester, preferably a p-nitrophenyl phosphonic ester.

7. Array according to claim 6, characterized in that the at least one compound is selected from the compounds with the formulas 1 to 23 listed in figures 2A, 2B and 2C.

8. Method of determining qualitatively and quantitatively the binding of a compound to an enzyme selected from the group consisting of lipase and esterase comprising the steps: a) providing

I) a protein array according to any one of claims 1 to 5 comprising said enzymes or

II) an array comprising at least one compound inhibiting or potentially inhibiting said enzyme bound to a solid support according to claim 6, b) contacting array I) with at least one compound inhibiting or potentially inhibiting said enzyme or contacting array II) with an enzyme forming an inhibitor/enzyme complex, and c) detecting binding of the at least one compound to said enzyme.

9. Method according to claim 8, characterized in that the at least one compound is an ester, preferably a p-nitrophenyl phos- phonic ester.

10. Method according to claim 8 or 9, characterized in that the at least one compound is selected from the compounds with the formulas 1 to 23 listed in figures 2A, 2B and 2C.

11. Method according to any one of claims 8 to 10, characterized in that the at least one compound is biotinylated.

12. Method according to any one of claims 8 to 11, characterized in that binding of the at least one compound to said enzyme is detected by contacting the array with a first antibody directed towards said enzyme or the at least one compound and optionally with a second antibody directed to said first antibody, wherein the first or second antibody is labeled with an enzyme, a fluorophore, a biotin, an enzyme substrate and any other affinity tag.

13. Method according to claim 12, characterized in that the enzyme is selected from the group consisting of horseradish perox- idase, alkaline phosphatase and luciferase.

14. Method according to claim 12, characterized in that the fluorophore is selected from the group consisting of cyanines, merocyanines, boradiazaindacene, rhodamins, anthracenes, ni- trobenzoxadiazoles and combinations thereof.

15. Kit for determining binding of a compound to an enzyme selected from the group consisting of lipase and esterase comprising:

- an array covered by a hydrogel as defined in any of claims 1 to 4,

- a lipase and/or an esterase and

- optionally means for detecting the binding of the compound to said enzyme.