WO2003060460A2

WO2003060460A2 - Multi-target screening

Info

Publication number: WO2003060460A2
Application number: PCT/HU2003/000001
Authority: WO
Inventors: Árpád FURKA
Original assignee: Furka Arpad
Priority date: 2002-01-03
Filing date: 2003-01-02
Publication date: 2003-07-24
Also published as: AU2003202103A8; AU2003202103A1; WO2003060460B1; WO2003060460A3

Abstract

Methods are disclosed for identifying natural macromolecules of pharmaceutical interest and their macromolecular or synthetic ligands. The targets and their ligands are identified in the same binding experiment executed with multi-component target libraries and multi-component ligand libraries. Libraries of natural macromolecules tethered to non-coded or coded solid supports are also claimed.

Description

MULTI-TARGET SCREENING

The present invention discloses a new method for identification of natural macromolecular target molecules such as proteins, nucleic acids, polysaccharides and their macro- molecular or synthetic ligands. According to the invention multi-component target libraries and multi-component ligand libraries are used in screening tests whereby the target molecules and their ligands are identified in the same experiment.

It is known in the art that bindings test are very extensively used in identification of new drug candidates. The screening methods applied in traditional pharmaceutical research are based on testing individual compounds. The split-mix method published by Furka et al. in 1998 and 1991 made it possible to prepare millions of compounds in a single experiment and methods were developed for replacing the sin- gle ligands in binding tests with multi-component libraries. The first such method was published by Lam et al. in 1991. In their binding experiment one target protein and a multi- component peptide library was used. Other methods like that of Houghten et al. (1991) and Campian et al. (1998) also used a single target.

The series of individual compounds prepared by parallel synthetic methods are usually also screened against a single protein target. Automatic machines were developed to carry out the assays. As a result of extensive automation the high throughput screening (HTS) methods made it possible to realize thousands of screening tests in one day.

As a result of the success of the human genome project tens of thousands of new protein target molecules became available. In an ideal situation every new synthetic drug candidate should be tested against all available targets. Considering, however, that the available synthetic methods enable their users to produce new compounds in almost unlimited number, by using the screening methods known to date it would be impossible to test all synthetic compounds against all potentially available targets. The Multi-Target Screening Method (MTSM) according to the present invention is designed to face this challenge.

The present invention comprises a method for identification of natural macromolecules and of their natural or syn- thetic ligands in the same binding experiment by a) executing said binding experiment using i) a library of natural macromolecules the components of which are tethered to the encoded or non-coded units of a multiparticle solid sup- port and incubating it with a dissolved library of labelled natural macromolecules, or ii) a library of natural macromolecules the components of which are tethered to the encoded or non-coded units of a multiparticle solid sup- port and incubating it with a dissolved library of labelled synthetic compounds, or iii) a library of synthetic compounds the components of which are tethered to the encoded or non- coded units of a multiparticle solid support and incubating it with a dissolved library of labelled natural macromolecules, whereby in the tethered libraries used for the binding experiment each component is attached to a different solid support unit b) selecting the units which show the presence of an attached component of the dissolved library and c) identifying the binding pairs of molecules by analyzing the content of the selected units. The natural macromolecule is preferably a protein, a nucleic acid or a polysaccharide. Both the tethered and the dissolved libraries consist of at least two components.

The solid support consists of a natural or synthetic polymer or of glass all of which may be functionalized; having the shape of beads, sheets, cubes, rods, polygons, or other geometric forms.

The solid support consists of units that are non-coded or encoded by their size or shape, or by colourless or colour molecular tags, or by fluorescent tags, or by embedded magnetic particles or by embedded transponders.

The components of the dissolved library are labelled preferably by colour or fluorescent tags or by radioactivity. The identity of the component of the dissolved library that binds to the component tethered to a selected unit is determined after its removal from the unit by mass spectro- etry, capillar electrophoresis, gas chromatography or other spectroscopic methods. The identity of the natural macromolecule or the synthetic compound tethered to a selected encoded unit is deduced after determining its code.

The identity of a peptide compound tethered to a selected non-encoded unit is determined by submitting the unit to a ino acid sequence determination.

The identity of a natural macromolecule or synthetic compound tethered to a selected non-encoded unit is determined by cleaving the natural macromolecule and synthetic compound from the unit and then analyzing the same by mass spectro- metry, HPLC, capillar electrophoresis, gas chromatography or other spectroscopic methods.

The components of synthetic compound libraries are tethered to solid support and are tested with solutions of libraries or mixtures of target molecules carrying colour, fluorescent or other kinds of labels. The components of target libraries are immobilized by attaching them to microscopic solid beads. The components of immobilized compound libraries on the other hand, are attached to the microscopic solid beads used as solid supports in their synthesis.

The screening process is expected to identify the binding pairs of molecules.

According to the first preferred embodiment of the present invention the components of immobilized target libraries are natural macromolecules, proteins or nucleic acids that are tethered to a solid carrier which is preferably in the form of small beads. It is suitable to attach each component of the library to a different bead. Preparation of such a protein library, for example, might be carried out in at least two steps:

• In one step immobilized protein samples are made by attaching each of the available protein to a different sample of resin;

• a further step consists in creating a sample of a multi-target library by combining in an isolated place preferably in a small tube one bead from each immobilized protein sample.

The solid carriers are preferably small units of func- tionalized natural or synthetic polymers or glass beads. The functional groups ensure the attachment sites in the beads. Previous encoding of the unit samples facilitates identification of the target molecule attached to a given bead. If non-encoded beads were used in construction of protein libraries, for example, the proteins themselves would have to be analysed later in the screening experiments in order to determine their identity.

The number of targets forming a library is not limited; it depends on the availability the targets . The capacity and shape of the beads has to be optimized according to the de- mands, specifically of the analytical methods applied. In many cases the protein tethered to a non-coded unit can be identified by determining the molecular weight of the protein cleaved from the unit using mass spectrometry (MS) . In other cases the protein is submitted to MS after enzymatic digestion (Yates, Methods in Enzymology Vol. 271, p. 351). In order to facilitate identification of the target proteins, one unit is removed from the tethered protein samples for analysis prior to construction of the immobilized target library. The proteins are then cleaved from the units, sepa- rately submitted to MS and their spectra are recorded.

The target proteins are attached to the units by either strong non-covalent bonds or by easily cleavable covalent bonds. One possibility for applying non-covalent bonds is to use biotinylated beads (or commercially available biotin beads) and the proteins in avidinated form or vice versa (Savage et al. Avidin-Biotin Chemistry, 1992) . Mixing the biotinylated units with the solution of an avidinated protein, the proteins are attached to the beads as a result of a strong non-covalent bond. Covalent immobilization methods that permit subsequent release of the proteins are also known (Horton and Swaisgood, Methods in Enzymology Vol. 135, p. 130) and can be used to carry out the present invention. Commercial resins carrying photocleavable linkers serve also well . According to a further preferred embodiment of the present invention the target libraries are immobilized on encoded units. According to this method the target macromolecules can indirectly be identified in binding tests preferably if they are attached to previously encoded beads. In this case each protein has to be attached to a unit carrying a different code. One possibility for encoding would be the use of units of different size and/or shape. Colour or fluorescence can also be applied to label the different units (Campian et al. 1994). Encoding by sequences (Brenner and Lerner 1992, Needels et al. 1993, Nielsen et al. 1993, Niko- laiev et al . 1991) or by molecular tags forming a binary encoding system (Ohlmeyer et al . 1993, Ni et al . 1996) is also a possibility. Application of commercially available units such as beads each containing an embedded transponder is also possible according to the invention.

When using encoded support, each target is coupled by covalent bond to a sample carrying a different code. The code of each bead sample and the name of the target attached to it is preferably recorded e.g. in a manner known per se. Then, when required, the protein can be identified by determining the code of the bead.

According to the present invention soluble mixtures of labelled macromolecules are formed by labelling them individually. Labeling is accomplished preferably by attaching to them a colour or fluorescent or radioactive tag. All molecules or a subgroup of them carry the same label. Before using the samples of labelled macromolecules are submitted to MS and their spectra are recorded.

Protein mixtures extracted from cells or other sources can also be labelled by colour, fluorescent or radioactive markers. In this case all molecules have the same label. The mixtures of labelled macromolecules are used in soluble form in binding experiments carried out with compound libraries or target libraries tethered to beads. According to yet another preferred embodiment of the present invention synthetic combinatorial compound libraries are created and used. Preparation of combinatorial compound libraries in both tethered and soluble form is described in prior art. In the synthesis of libraries tethered to non- encoded beads it is preferable to follow the split-mix method of Furka et al . (1991). Tethered libraries on encoded beads may be synthesized according to Brenner and Lerner (1992), Needels et al . (1993), Nielsen et al. (1993), Nikolaiev et al. (1991), Ohlmeyer et al . (1993) or Ni et al . (1996). Preparation of soluble labelled libraries was published by Sebestyen et al. in 1998. The methods for preparing compounds labelled with radioactive isotopes are also well developed and can be used for the present invention. Another possibility is the use of beads with embedded transponders in the synthesis as described by Mandecki (2002) .

In the following some of the steps used for accomplishing the method according to the present invention are disclosed in more details.

In binding experiments always at least two different li- braries are used: one in tethered form and the other one in dissolved form.

The components of the dissolved library always carry a label (marker) .

The units (beads) comprising the tethered library are incubated with the other library dissolved in an appropriate aqueous solvent such as a buffer. After the incubation period the solution is removed by filtration then the units are washed. Depending on the type of marker used to label the dissolved library components, the pool of beads is examined for colour, fluorescent or radioactive beads. Appearance of such beads is an indication of binding of components of the dissolved library to components of the tethered library. These distinguished beads are removed and submitted to analysis in order to identify the binding pairs of molecules. Identification of the binding pairs of molecules is brought about in the following manner according to a preferred embodiment of the invention: The distinguished solid units are separately eluted with a buffer. The eluant is used in determination of the component of the dissolved library that participated in binding. The solid unit, on the other hand, makes possible to determine the tethered partner of binding.

Determination of macromolecules in the eluant may be carried out as follows : If the dissolved library is constructed by mixing to- gether known proteins, determination of molecular weight is most often sufficient for identification of the protein participating in binding. If the dissolved library is a mixture of unknown proteins (such as a tissue extract) then prior to the mass spectroscopic determination the eluted protein is submitted to fragmentation by known methods.

Determination of the synthetic ligand in the eluant may be accomplished in the following manner: A dissolved synthetic library is used in the binding experiments and deter- mination of the molecular weight is carried out by MS. This method is sufficient to identify the eluted compound in most cases .

In some cases determination of the molecular weight identifies a group of compounds that have the same molecular weight. Such compounds have to be individually synthesized, then submitted separately to binding experiments each with its acromolecular binding partner. The right ligand is unequivocally identified by these experiments.

Identification of the macromolecules remaining on the distinguished beads after elution may be accomplished using different methods:

The macromolecule tethered to an encoded bead can be identified by determining the code of the bead. This is done by examining the colour or fluorescence of the bead, or by reading de code written in the embedded transponder (Mandecki 2002) .

If the beads are encoded by molecular tags, these tags are cleaved from the beads then determined by known methods (Ohlmeyer et al . 1993, Ni et al . 1996) . If the beads are non-coded, prior to further analysis the macromolecules are cleaved from the beads. Macromolecules attached to the beads by avidin-biotin binding are eluted from the beads by using an appropriate buffer.

If the macromolecules are immobilized by covalent bond then the bead is submitted to irradiation or to an appropri- ate chemical reagent that cleaves the bond. The macromolecules are then eluted from the bead and submitted to MS to determine the molecular weight.

If the molecular weight does not unequivocally identify the macromolecule (e. g. more than one macromolecule has the same molecular weight) then, after a fragmentation, a second MS determination is carried out.

Identification of synthetic compounds tethered to distinguished beads may be carried out by way of different ac- complishments :

If the bead remaining after elution belongs to a tethered peptide library, then the amino acid sequence of the tethered peptide is determined by directly submitting the bead to micro sequencing (Lam et al. 1991) . If the distin- guished bead belongs to an organic compound library synthesized on non-encoded beads, the compound is cleaved from the bead and in order to identify it, its molecular weight is determined by MS. In cases where more than one compound has the determined molecular weight, these compounds are separately re-synthesized and the binding test is repeated with them using the already determined macromolecular binding partner in the experiments .

Where the distinguished bead is. encoded - the code of the bead has to be determined in order to identify the teth- ered compound. This is done by examining the colour or fluorescence of the bead or by reading the code written in the embedded transponder (Mandecki 2002) . If the beads are encoded by molecular tags, these tags are cleaved from the beads and are then determined by known methods (Ohlmeyer et al. 1993, Ni et al . 1996).

Further subjects of the invention are the new products comprising libraries of natural macromolecules the components of which are tethered to the non-coded or encoded units of a multiparticle solid support whereby each component of the li- brary is attached to a different unit. Specifically important products according to the invention comprise libraries where the macromolecules are proteins, nucleic acids and/or polisaccharides . The solid supports consisting of natural polymers, synthetic polymers and/or glass. The mulitparticle units may be functionalized to serve as attachment sites of immobilization of the targets. The shape of the units may vary including beads, sheets, cubes, rods, polygons, or other geometric forms.

The Multi-Target Screening Method using the multiparti- cle solids comprising the above specified components in tethered form have several advantages as compared with the methods known in prior art. The method is much more efficient than the conventional single target screening assays since hundreds or even thousands of experiments can be replaced by a single one. For the same reason the costs of multi-target screening are less if compared with the costs of the conventional methods. Application of the multi-target screening will help to speed up determination of interactions between macromolecules as well as identification of new targets and new drug leads.

The following examples show more details of the invention without the intention of limitation.

Examples Example 1. Screening of immobilized protein library with synthetic tripeptide amide library

Preparation of immobilized protein library. 0.5 mg of thionyl chloride activated succinamidopropyl glass beads prepared according to Horton and Swaisgood (Methods in Enzymol- ogy Vol. 135, p. 131) were shaken at room temperature for 3 hours with 0.2 mg protein dissolved in 0.1 ml of aqueous phosphate buffer at pH 7 then washed 5 times with the same buffer. Following this procedure, the immobilized samples of the following proteins were prepared: lysozyme, myoglobin, bovine serum albumin, α-lactoglobulin, LH-RH antibody, oval- bumin. The immobilized protein library was constructed by transferring into a small tube one bead from each immobilized protein sample.

Mass spectra of the immobilized proteins . One bead from each immobilized protein sample was separately transferred to a small tube then 1 M hydroxylamine in 0.1 ml of 50 mM sodium phosphate buffer was added to each tube and kept at room temperature for 3 hours with occasional shaking. The cleaved proteins were submitted to electronspray ionization mass spectrometry (ESI MS) in order to determine the molecular weight and the spectra were recorded.

Synthesis of a labelled tripeptide amide library. A tripeptide amide library was synthesized on Rink resin using 19 a ino acids (cystein was omitted) and the Fmoc strategy and following the split-mix method of Furka et al. After the third coupling step the peptides were acylated with Dansyl- chloride then were cleaved from the resin as peptide amides and stored in lyophilized form.

Binding assay using an immobilized protein library and a labelled tripeptide library. The protein library was washed with 3x0.1 ml of pH 7 phosphate buffer containing 0.1 mg un- labeled Gly-Ala-Ala-NH₂ then with 3x0.1 ml of the same buffer. 0.5 mg of the labelled tripeptide amide library was dissolved in 0.1 ml pH 7 aqueous phosphate buffer then added to the i - mobilized protein library. After incubation at room temperature for 4 hours the solution was removed then the beads were washed 3x0.2 ml of the same buffer. The bead that showed fluorescence when lighted with UN lamp was removed for analysis. In order to remove the attached peptide the fluorescent bead was eluted by occasional shaking with 0.1 ml pH 2 aqueous buffer (2% formic acid, 8% acetic acid) for 3 hours then it was washed with 3x0.2 ml of the same buffer and 3x0.2 ml distilled water. Then the protein was cleaved from the bead by adding 0.1 ml pH 7 aqueous phosphate buffer containing 1 M hydroxylamine and kept at room temperature for 3 hours with occasional shaking.

Identification of the binding protein-tripeptide pair. The protein sample cleaved from the bead was submitted to ESI MS and the result compared to the previously recorded MS spectrums of proteins. This revealed that the protein binding to the tripeptide is the LH-RH antibody. The peptide eluted from the bead was also submitted to ESI MS. From the determined molecular weight and of the calculated molecular weights of the 6859 components of the labelled peptide li- brary it could be deduced that the tripeptide participating in binding is one of the following Dansylated tripeptide amides: PGR, GRP, GPR, RGP, RPG and PRG. These tripeptides were separately synthesized then submitted to binding test with one bead of the immobilized LH-RH antibody. The Dansylated RPG amide showed binding. This identified the binding pairs of molecules as LH-RH antibody and its ligand Arg-Pro-Gly-ΝH₂.

Example 2.

Screening of a tethered synthetic tripeptide library with a library of proteins Preparation of the library of biotinylated protein.

0.1 mg portions of lysozyme, myoglobin, α-lactoglobulin, LH-RH antibody, ovalbumin were biotinylated following the description of the manufacturer of the reagent. Thereafter the molecular weight of each protein was determined by ESI-MS and the spectra were recorded. Additional 0.1 mg portions of the same proteins were mixed and then biotinylated.

Synthesis of a tethered tripeptide amide library. In the synthesis of the tripeptide library, Tentagel S NH₂ resin was used as solid support and the Fmoc strategy and the split-mix method were followed. First Fmoc-GABA (Fmoc-α-aminobutyric acid) was coupled to the resin then the tripeptide library was constructed using 19 Fmoc-amino acids (cystein was omitted) . Finally the protecting groups were removed. Enzyme-linked binding assay using the immobilized tripeptide library and the library of biotinylated proteins . The assay was carried out as described by Chen et al . (Methods in Enzymology Vol. 267 p. 221) with the following exceptions: (i) the single biotinylated protein was replaced by the prepared 5 member library of biotinylated proteins; (ii) the molecular weight of the biotinylated protein eluted from the bead was determined by ESI-MS and the binding protein was identified as the LH-RH antibody. Microsequencing of the tripeptide tethered to the bead yielded: Arg-Pro-Gly.

Claims

1. Multi-target screening method for identification of natural macromolecules and of their natural or synthetic ligands in the same binding experiment by a) executing said binding experiment using i) a library of natural macromolecules the components of which are tethered to the encoded or non-coded units of a multiparticle solid support and incubating it with a dis- solved library of labelled natural macromolecules, or ii) a library of natural macromolecules the components of which are tethered to the encoded or non-coded units of a multiparticle solid support and incubating it with a dissolved library of labelled synthetic compounds, or iii) a library of synthetic compounds the components of which are tethered to the encoded or non-coded units of a multiparticle solid support and incubating it with a dissolved library of labelled natural macromolecules, whereby in the tethered libraries used for the binding experiment each component is attached to a different solid support unit b) selecting the units which show the presence of an attached component of the dissolved library c) identifying the binding pairs of molecules by analyzing the content of the selected units.

2. The method of claim 1 wherein the natural macromolecules are proteins, nucleic acids or polysaccha- rides .

3. The method of claim 1 or 2 wherein both the tethered and the dissolved libraries consist of at least two components .

4. The method of any of claims 1 to 3 wherein the solid support consists of a natural or synthetic polymer or of glass all of which may be functionalized; having the shape of beads, sheets, cubes, rods, polygons, or other geometric forms.

5. The method of any of claims 1 to 4 wherein the solid support consists of units that are non-coded or encoded by their size or shape, or by colourless or colour molecular tags, or by fluorescent tags, or by embed- ded magnetic particles or by embedded transponders.

6. The method of any of claims 1 to 5 wherein the components of the dissolved library are labelled by colour or fluorescent tags or by radioactivity.

7. The method of any of claims 1 to 6 wherein the identity of the component of the dissolved library that binds to the component tethered to a selected unit is determined after its removal from the unit by mass spec- tro etry, HPLC, gas chromatography, capillar electrophoresis or other spectroscopic methods.

8. The method of any of claims 1 to 7 wherein the identity of the natural macromolecule or the synthetic compound tethered to a selected encoded unit is deduced after determining its code.

9. The method of any of claims 1 to 8 wherein the identity of a peptide compound tethered to a selected non-encoded unit is determined by submitting the unit to amino acid sequence determination.

10. The method of any of claims 1 to 9 wherein the identity of a member of the group natural macromolecule or synthetic compound tethered to a selected non-encoded unit is determined by cleaving the natural macromolecule or synthetic compound from the unit and then analyzing the same by mass spectrometry, HPLC, capillar electrophoresis, gas chromatography or other spectroscopic methods.

11. A library of natural macromolecules the components of which are tethered to the non-coded or encoded units of a multiparticle solid support whereby each com- ponent of the library is attached to a different unit.

12. A library according to claim 11 where the macromolecules are proteins, nucleic acids and/or polisac- charides .

13. A library according to claim 11 or 12 wherein the solid support consists of natural or synthetic polymer or of glass all of which may be functionalized; having the shape of beads, sheets, cubes, rods, polygons, or other geometric forms.