WO2003091450A1

WO2003091450A1 - Method for evaluating a therapeutic potential of a chemical entity

Info

Publication number: WO2003091450A1
Application number: PCT/DK2003/000256
Authority: WO
Inventors: Jens BITSCH-NØRHAVE; Rene Hummel; Jens Damsgaard Mikkelsen
Original assignee: Azign Bioscience AS
Current assignee: Azign Bioscience AS
Priority date: 2002-04-24
Filing date: 2003-04-15
Publication date: 2003-11-06
Anticipated expiration: 2004-10-24
Also published as: AU2003226947A1

Abstract

The invention relates to methods for predicting a therapeutic potential of a chemical entity and a specific differential display array.

Description

METHOD FOR EVALUATING A THERAPEUTIC POTENTIAL OF A

CHEMICAL ENTITY

FIELD OF INVENTION

BACKGROUND OF INVENTION Gene expression profiles have become increasingly important tools in the biotechnology industry and related fields. By clustering the data generated by gene expression, important gene-drug relationships can be established. One of the applications is differential gene expression, where expression of genes in different cells or tissues (normally a control sample and a sample of the cell or tissue of interest) is compared, and any difference in the mRNA expression profile is determined.

In gene expression analysis using arrays, an array of "probe" nucleotides is contacted with a nucleic acid sample of interest such as mRNA concerted into cDNA from a particular tissue or cell. Contact is carried out under hybridisation conditions favourable for hybridisation of nucleic acids complementary to the "probe" nucleotides on the array. Unbound nucleic acid is then removed by washing. The resulting pattern of hybridised nucleic acid provides information regarding the gene expression profile of the sample tested on the array. Gene expression analysis is used in a variety of applications including identification of novel expression of genes, correlation of gene expression to a particular tissue or a particular disease, identifying effects of agents on the cellular expression such as in toxicity testing and in identifying drugs.

WO 01/51667 (Integriderm) describes informative nucleic arrays and methods for making same. Various arrays are described in Yamada, M et al, Neuroscience Letters, (2001) 301 : 183-186; and Yamada, M et al, Biochemical and Biophysical Research Communications (2000) 278: 150-157. Further, various aspects of gene expression profiling is described in Landgrebe, J et al, Journal of Psychiatric Research (2002) 36: 119-129; and Scherf, U et al, Nature Genetics (2000) 24: 236-244.

The present invention provides methods for evaluating a therapeutic potential of a chemical entity by comparing the gene expression profile of the chemical entity with gene expression profiles of a plurality of reference compounds.

BRIEF DISCLOSURE OF THE INVENTION

Accordingly, in a first aspect the invention relates to a method for evaluating a therapeutic potential of a chemical entity comprising: • generating gene expression profiles of a plurality of reference compounds, wherein:

^■ each reference compound has a known biological effect, and

^■ each reference compound may be grouped in one of at least two distinct classes according to its biological effect, and

^■ each distinct class has at least one reference compound as a member;

• generating a gene expression profile of the chemical entity;

• comparing the gene expression profile of the chemical entity with the gene expression profiles of the reference compounds;

• predicting the therapeutic potential of the chemical entity on the basis of said comparison.

In a further aspect, the invention relates to a specific differential display array for generating a gene expression profile of a test compound comprising: a plurality of polynucleotide spots associated with a surface of a support, wherein the individual polynucleotide spot comprises: a individual gene fragment selected with the purpose to contribute to a optimal differentiation between one of two distinct classes of therapeutic action of the test compound or one of two distinct classes according to action on a target of the test compound.

In still further aspects, the invention relates to a database comprising a plurality of gene fragments and a chemical entity evaluated by the method of above.

Further objects of the invention will be apparent to the person skilled in the art from the following detailed description and examples.

DETAILED DESCRIPTION OF THE INVENTION

In a first aspect, the present invention provides a method for evaluating a therapeutic potential of a chemical entity comprising:

• generating gene expression profiles of a plurality of reference compounds, wherein:

^■ each reference compound has a known biological effect, and

^■ each distinct class has at least one reference compound as a member;

• generating a gene expression profile of the chemical entity;

• comparing the gene expression profile of the chemical entity with the gene expression profiles of the reference compounds; • predicting the therapeutic potential of the chemical entity on the basis of said comparison.

In one embodiment, the generation of an expression profile of each of the reference compounds comprises: • obtaining a first expression profile of a biological material not treated with the reference compound;

• obtaining a second expression profile of a biological material treated with the reference compound;

• generating an expression profile of the reference compound based on differences between the first expression profile and the second expression profile. In a second embodiment, the generation of an expression profile of the chemical entity comprises:

• obtaining a first expression profile of a biological material not treated with the chemical entity;

• obtaining a second expression profile of a biological material treated with the chemical entity;

• generating an expression profile of the chemical entity based on differences between the first expression profile and the second expression profile.

In a third embodiment, each reference compound may be grouped in one of two distinct classes according to its biological effect. In a further embodiment, each reference compound may be grouped in one of three distinct classes according to its biological effect. In a still further embodiment, each reference compound may be grouped in one of four distinct classes according to its biological effect.

In a further embodiment, each distinct class has at least two, three, four or five reference compounds as members.

In a special embodiment, each reference compound may be grouped in one of the two classes of antidepressants and antipsychotics. In a further special embodiment, each reference compound may be grouped in one of the two classes of a therapeutic action with one or more side effects and a therapeutic action without one or more side effects. In a more special embodiment, the therapeutic action is antidepressant and the side effect is nausea, dizziness, anxiety, or drowsiness. In a further embodiment, each reference compound may be grouped in one of two distinct classes according to its action on a target.

In a special embodiment, each reference compound may be grouped in one of the two classes of 5-HT1 receptor agonists and 5-HT2 receptor agonists. In a further embodiment of the invention, the gene expression profiles of the plurality of reference compounds and the gene expression profile of the chemical entity are generated by the use of a differential display technique. In a special embodiment, the differential display technique used is restriction fragment differential display. In a further special embodiment, the differential display technique used is the restriction fragment differential display (RFDD) PCR analysis as described in US Patent No. 6,261 ,770 (Azign Bioscience).

In a further embodiment, the restriction fragment differential display technique is performed by the use of a specific differential display array comprising: a plurality of polynucleotide spots associated with a surface of a support, wherein the individual polynucleotide spot comprises: an individual gene fragment selected with the purpose to contribute to a optimal differentiation between one of the at least two distinct classes of biological effect.

In a special embodiment the individual gene fragments are selected by the use of mathematical procedures such as Gene Shaving, Principal Component Analysis (PCA), or statistical analysis.

In a further embodiment, the individual gene fragments is selected by generating gene expression profiles of a plurality of array reference compounds, wherein: • each array reference compound has a known biological effect, and

• each array reference compound may be grouped in one of at least two distinct classes according to their biological effect, and

• each distinct class has at least one array reference compound as a member. In a still further embodiment, the group of array reference compounds and the group of reference compounds have identical members.

In a further aspect, the present invention provides a specific differential display array for generating a gene expression profile of a test compound comprising: a plurality of polynucleotide spots associated with a surface of a support, wherein the individual polynucleotide spot comprises an individual gene fragment selected with the purpose to contribute to a optimal differentiation between one of at least two distinct classes of biological effect of the test compound.

In one embodiment, the individual gene fragments are selected by the use of mathematical procedures such as Gene Shaving, Principal Component Analysis (PCA), or statistical analysis.

In a second embodiment, the individual gene fragments are selected by generating gene expression profiles of a plurality, of array reference compounds, each array reference compound being grouped in one of the at least two distinct classes according to their biological effect. In a further embodiment, the generation of a gene expression profile of each of the array reference compounds comprises:

• obtaining a first expression profile of a biological source not treated with the array reference compound; • obtaining a second expression profile of a biological source treated with the array reference compound;

• generating an expression profile of the array reference compound based on differences between the first expression profile and the second expression profile. In a still further embodiment, each array reference compound may be grouped in one of two distinct classes according to its biological effect. In a further embodiment, each array reference compound may be grouped in one of three distinct classes according to its biological effect. In a still further embodiment, each array reference compound may be grouped in one of four distinct classes according to its biological effect.

In a special embodiment, each array reference compound may be grouped in one of the two classes of antidepressants and antipsychotics.

In a further embodiment, each array reference compound may be grouped in one of the two classes of a therapeutic action with one or more side effects and a therapeutic action without one or more side effects.

In a special embodiment, the therapeutic action is antidepressant and the side effect is nausea, dizziness, anxiety, or drowsiness.

In a further embodiment, each array reference compound may be grouped in one of two distinct classes according to its action on a target. In a special embodiment, each array reference compound may be grouped in one of the two classes of 5-HT1 receptor agonists and 5-HT2 receptor agonists.

In a still further embodiment, the gene expression profiles of the plurality of array reference compounds are generated by the use of a differential display technique. In a further embodiment of the specific differential display array, the individual polynucleotide spot is in single stranded or double stranded form.

In a still further embodiment, the individual polynucleotide spot is of DNA, RNA, cDNA, natural, synthetic, semisynthetic origin or is a chemical analogous such as LNA and PNA. In a further embodiment, the support is a porous support. In a still further embodiment, the support is a solid support.

In a further embodiment, the solid support is made of a flexible or rigid material.

In a still further embodiment, the array comprises from 4 to 40,000 polynucleotide spots. In a further embodiment, the array comprises two or more spots comprising a gene fragment from the same expressed gene. In a special embodiment, the spots comprising a gene fragment from the same expressed gene comprise the same gene fragment. In a further special embodiment, the spots comprising a gene fragment from the same expressed gene comprise different gene fragments from the same gene. In a still further special embodiment thereof, all said fragments are chosen in such a way that they are non-overlapping regions. The number of gene fragments to be chosen is dependent inter alia on the length of the expressed gene. The use of two or more gene fragments from the same expressed gene ensures an improved specificity of the array. Furthermore, the use of two or more gene fragments from the same expressed gene will allow a statistical evaluation of the signal from the individual gene fragment.

In a still further aspect, the present invention provides a database comprising a plurality of gene fragments being present on the specific differential display array as described above. In a special embodiment, the database further comprises additional expressed genes that are substantially relevant to the biological state.

In a further aspect, the present invention provides a chemical entity evaluated by the method as described above.

In a further embodiment, the biological material is selected from the group consisting of tissues, organs, biological fluids or parts or combinations thereof. In a special embodiment, the biological material and the biological source are of the same kind, i.e. of the same kind of origin, such as coming from the same type of tissue, organ, biological fluid or part or combinations thereof, or of the same organism or the same type of organism or the same cell type etc.

In a further embodiment, the individual gene fragment has a length of from 5 to 1000 nucleotides, such as from 50 to 1000 nucleotides. In a special embodiment, the identified fragment has a length of from 5 to 750 nucleotides, such as from 50 to 750 nucleotides. In a further special embodiment, the identified fragment has a length of from 20 to 100 nucleotides. In a still further embodiment, the identified fragment has a length of from 40 to 500 nucleotides. In a special embodiment, the individual gene fragment is produced by one or more primers specific for said gene fragment.

In a further embodiment of the methods for generating an expression of the reference compound or the array reference compound, each biological material or source is labelled with a unique label (e.g. Cy3 and Cy5 for each sample, respectively).

Definitions

In the context of the present application and invention the following definitions apply: The term "polynucleotide" is intended to mean a single or double stranded polymer composed of nucleotides, e.g. deoxyribonucleotides and/or ribonucleotides from about 5 to about 9,000 nucleotides in length, such as from about 10 to about 9,000, from about 5 to about 6,000, from about 10 to about 6,000, from about 5 to about 3,000, from about 10 to about 3,000, from about 5 to about 1 ,500, from about 10 to about 1 ,500, from about 5 to about 1,000, from about 10 to about 1 ,000, from about 50 to about 1 ,000, or from about 50 to about 750.

The term "complementary" or "complementarity" is used in relation to the base- pairing rules of nucleotides well known for a person skilled in the art. Polynucleotides may be complete or partial complementary. Partial complementarity means that at least one nucleic acid base is not matched according to the base pairing rules. Complete complementarity means that all nucleotides in a polynucleotide match according to the base pairing rules. The degree of complementary between polynucleotides affects the strength of hybridisation between two polynucleotide strands. The inhibition by hybridisation of the complementary polynucleotide to the target polynucleotide may be analysed by techniques well known for a person skilled in the art, such as Southern blot, Northern blot, and the like under conditions of high stringency. A partially (substantially) homologous polynucleotide will compete for and inhibit the binding of a completely homologous sequence to the target sequence under low stringency.

The term "homology" is intended to mean the degree of identity of one polynucleotide to another polynucleotide. According to the invention the term homology is used in connection with complementarity between polynucleotides within a family or between species. There may be complete homology (i.e. 100% identity) between two or more polynucleotides. The degree of homology may be determined by any method well known for a person skilled in the art.

The term "polynucleotide composition" is intended to mean a composition comprising a polynucleotide together with an excipient. The polynucleotide compositions are applied as spots on the array. The term "polynucleotide composition" includes also control or calibrating compositions such as, e.g. compositions comprising polynucleotides corresponding to housekeeping genes.

The terms "expression profile", "differential expression profile" or "gene expression profile" are intended to mean the expression of the mRNAs in a biological material. While an expression profile encompasses a representation of the expression level of at least one mRNA, in practice the typical expression profile represents the expression of several mRNAs. For example, an expression profile used according to the present invention represents the expression levels of at least from 1 to 50,000 or more different mRNAs in a biological material. The expression level of the different mRNAs is the same or different. The expression of mRNAs may be up- or down regulated resulting in different expression profiles.

The term "biological material" include within its meaning organisms, organs, tissues, cells or biological material produced by a cell culture. The biological material may be living or dead. The material may correspond to one or more cells from the organisms, in case the organism is a multicellular organism, the biological material may correspond to one or more cells from one or more tissues creating the multicellular organism. The biological material to be used according to the invention may be derived from particular organs or tissues of the multicellular organism, or from isolated cells obtained from a single or multicellular organism.

The term "biological source" include within its meaning organisms, organs, tissues, cells or biological material produced by a cell culture. The biological material may be living or dead. The material may correspond to one or more cells from the organisms, in case the organism is a multicellular organism, the biological material may correspond to one or more cells from one or more tissues creating the multicellular organism. The biological material to be used according to the invention may be derived from particular organs or tissues of the multicellular organism, or from isolated cells obtained from a single or multicellular organism.

In obtaining the sample of nucleotides to be analysed from the biological material or biological source from which it is derived, the biological material or biological source may be subject to a number of different processing steps. Such steps might include tissue homogenisation, cell isolation and cytoplasma extraction, nucleic acid extraction and the like and such processing steps are generally well known for a person skilled in the art. Methods of isolating RNA from cells, tissues, organs or whole organisms are known to those skilled in the art and are described in Maniatis et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbour Press) (1989).

The term "organism" is intended to mean any single cell organism such as yeast or multicellular organism, including plants, fungi and animals, preferably mammals, such as humans, rats, mice, pigs, cows, horses, dogs, guinea pigs, ferrets, rabbits, sheep, apes, monkeys and cats.

The term "target polynucleotides" is intended to mean polynucleotides present in the biological material of interest. If the target polynucleotide has a complementary polynucleotide present on the specific differential display array, it will hybridise thereto and thus give rise to a detectable signal. The term "non-overlapping" is intended to mean that when the specific fragments used in the polynucleotide composition spots are obtained from the same polynucleotide, the regions are obtained from different parts of the polynucleotide and the different parts are located in such a manner that the regions not even overlap each other by a single nucleotide. In a polynucleotide of e.g. 1 ,000 nucleotides the regions 1-500 and 501-900 are non-overlapping. The non-overlapping polynucleotide regions may be located with a distance of one or more nucleotides from each other. The term "primer" is intented to mean a polymer of 3-50 nucleotides. The term "set of primers" is intended to mean one or more primers having the ability to amplify a polynucleotide region under suitable conditions. The length of the primers may be the same or different and dependent on the character of the polynucleotide region to be amplified. Design of such a set of primers is well known for a person skilled in the art. The set of primers having a sufficient length to specifically hybridise to a distinct polynucleotide in the sample and the length of the primers will be from about 3 to 50 nucleotides.

Methods of comparing the homology between different polynucleotides and/or parts of different polynucleotides are well known in the art. The polynucleotides may either belong to the same family or different families and/or being polynucleotides encoding the same polypeptide from the same or different species. Optimal alignment of nucleotides of a polynucleotide for comparison of the homologies may be conducted using the homology algorithm (Smith and Waterman, Adv. Appl. Math. 2: 482 (1981)), by the homology alignment algorithm of Needleman and Wunsch, J. Mol. Biol. 48:443 (1970), by the search for similarity method of Pearson and Lipman, Proc. Natl. Acad. Sci., USA 85: 2444 (1988), by computerised implementations of these algorithms by using for example CLUSTAL in the PC/Gene program by Intelligenetics, Mountain View, California, GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group (GCG) 575 Science Dr., Madison, Wisconsin, USA. The above-mentioned algorithms and computer implementations should be regarded as examples and the invention not limited thereto. The identification of a gene that is differentially expressed in the first and second biological source may be performed by a number of ways known by a person skilled in the art. Examples are the use of a differential display technique, such as restriction fragment differential display, northern blot analysis and microarray analysis.

Specific differential display array

A specific differential display array according to the invention has a multiplicity of individual polynucleotide spots, stably associated with a surface of a support. The specific differential display array comprises spots comprising a polynucleotide composition, wherein the polynucleotide regions within the composition are of known identity, usually of known sequence, as described later on in detail. The polynucleotide spots may be of convenient shape but most often circular, oval or any other suitable shape. The polynucleotide spots may be arranged in any convenient pattern across the surface of the support, such as in row or columns to form a grid, in a circular pattern and the like. Preferably the pattern of polynucleotide spots is arranged as a grid to facilitate the evaluation of the results obtained from the analyses in which the specific differential display array is used.

The specific differential display array according to the invention may be of a flexible or rigid solid support and the polynucleotide spots are stably associated thereto. By stably associated is meant that the polynucleotide spots will be associated in their position on the support during the analysis in which the specific differential display array is used, such as during different hybridisation, washing and detection conditions. The polynucleotide regions contained in the spots may be covalently or non-covalently associated to the surface of the support. Methods how to covalently or non-covalently bind the polynucleotide regions to the surface of the support are well known for a person skilled in the art and may be found in Ausubel et al., Current protocols in Molecular Biology, Greene Publishing Co. NY, (1995).

The support to which the individual polynucleotide spots are stably associated to is made of a flexible or rigid material. By flexible is meant that the support is capable of being bent or folded without breakage. By rigid is meant that the support is solid and does not readily bend, i.e. the support is not flexible. The support may be fabricated from a variety of materials, including plastics, ceramics, metals, gels, nitrocellulose, nylon, glass and the like.

The array may be produced according to any convenient methodology, such as preparing or obtaining the polynucleotides and then stably associate them with the surface of the support or growing them directly on the support. A number of array configurations and methods for their production are known to those skilled in the art and disclosed in US Patents Nos.: 5,445,934, 5,532,128, 5,556,752, 5,242,974, 5,384,261 , 5,405,783, 5,412,087, 5,424,186, 5,429,807, 5,436,327, 5,472,672, 5,527,681 , 5,529,756, 5,545,531 , 5,554,501 , 5,561 ,071 , 5,571 ,639, 5,593,839, 5,599,695, 5,624,711 , 5,658,734 and 5,700,637.

The support of the invention may have several configurations ranging from a simple to a more complex configuration depending on the intended use of the specific differential display array. The size and thickness of the specific differential display array is not critical as long as the array will function in the expected way and as long as the results obtained after use of the array are not changed. The number and amount of the polynucleotide spots is dependent on the intended use of the arrays as well as the detection system use to determine the expression profile of the biological material being evaluated by the aid of the specific differential display array. The number of the polynucleotide spots may vary from about 2 to about 100,000 such as, e.g., from about 2 to about 50,000, from about 10 to about 25,000, from about 100 to about 10,000, from about 100 to about 5,000, from about 100 to about 1 ,000, from about 400 to about 600 or about 500 polynucleotide spots, or at least 2 such as, e.g. at least 10, at least 25, at least 50, at least 100, at least 300, at least 400, at least 500 or at least 600 spots, or even more than 100,000 spots. The limitations of the number of the polynucleotide spots are dependent on the way in which the evaluation of the expression profile of the biological material is performed. The amount of the polynucleotide regions present in the polynucleotide spot may vary and the amount will be sufficient to provide adequate hybridisation and detection of the target nucleic acid. Generally the polynucleotides will be present in each spot at a concentration corresponding to an amount of 1 pg - 100 μg or less than 100 μg of the polynucleotide. Normally, only 1 polynucleotide region is present in each spot.

The copy number of the polynucleotide present in each polynucleotide spot will be sufficient to provide enough hybridisation for a target nucleic acid to yield a detectable signal, and generally range from about 1 nmol, 1 pmol, 500 fmol, 100 fmol, 50 fmol or less.

Other polynucleotide spots (control spots), which may be present on the specific differential display array, include spots comprising genomic DNA, housekeeping genes, negative and positive control polynucleotides and the like. These polynucleotide spots comprise polynucleotides, which are not unique, i.e they are not polynucleotide regions corresponding to differentially expressed genes. They are used for calibration or as control polynucleotides, and the function of these polynucleotide spots are not to give information of the expression of these polynucleotides, but rather to provide useful information, such as background or basal level of expression to verify that the analysis and the expression profiles obtained are relevant or not. Furthermore these control spots may serve as orientation spots.

The polynucleotide composition also comprises an excipient. Suitable excipients are solvents like e.g. water or any other aqueous medium, pH adjusting agents like buffering agents, stabilising agents, hybridising agents, coloring agents, labelling agents and the like. In general, the excipients used are inert, i.e. they do not have any polynucleotide related effect.

Method of preparing the specific differential display array The specific differential display array may be prepared (produced) using any convenient method and several methods are well known for a person skilled in the art, such as standard procedures according to Sambrook et al., (Molecular cloning: A laboratory manual 2^nd edition. Cold Spring Harbour Laboratory Press, New York.).

One means of preparing the array is: i) synthesising or otherwise obtaining the above mentioned gene fragments, ii) preparing the polynucleotide compositions to be used in each spot and then iii) depositing in the form of spots the polynucleotide compositions comprising the gene fragment onto the surface of the support. The gene fragments may be of DNA, RNA, cDNA, natural, synthetic, semisynthetic origin or chemical analogous such as LNA or PNA. The gene fragments may be obtained from any biological material such as, e.g., tissues or cells and/or produced by a cell culture. The biological material may be an organism, such as a microorganism, plant, fungus (e.g. yeast or mushrooms) or animal.

The gene fragments may be prepared using any conventional methodology such as automated solid phase synthesis protocols, PCR using one or more primers specific for the gene fragments and the like. In general, PCR is advantageous in view of the large numbers of gene fragments that must be generated for specific differential display array. The amplified gene fragments may further be cloned in any suitable plasmid vector to enable multiplication and storage of the amplified gene fragments. The prepared gene fragments may be spotted onto the support using any convenient methodology, including manual and automated techniques, e.g. by micro- pipette, ink jet pins etc. and any other suitable automated systems. An example of an automated system is the automated spotting device, Lucidea (Amersham Biosciences). The ready arrays may then be stored at suitable conditions until use.

Methods for the determination of expression profiles

Determination of expression profiles typically means determination of the expression level of multiple mRNAs, all of them corresponding to the spotted gene fragments on the array. The detection limit of the expression level of a mRNA may be approximately 0.2 ng or less of total RNA of the biological material used to hybridise each polynucleotide spot.

The expression profiles can be produced by any means known in the art, including but not limited to the methods disclosed by: Liang et al., (1992) Science 257: 967-971 ; Ivanova et al., (1995) Nucleic Acids Res 23: 2954-2958; Guilfoyl et al., (1997) Nucleic Acids Res 25(9): 1854-1858; Chee et al., (1996) Science 274: 610-614; Velculescu et al., (1995) Science 270: 484-487; Fiscker et al., (1995) Proc Natl Acad Sci USA 92(12): 5331.5335; and Kato (1995) Nucleic Acids Res 23(18): 3685-3690.

The method for the determination of an expression profile in a biological material or in a mixture of biological materials comprises obtaining a polynucleotide from the biological material(s), labelling said polynucleotide to obtain a labelled target polynucleotide sample, contacting at least one labelled target polynucleotide sample with an array as defined above under conditions which are sufficient to produce a hybridisation pattern and detecting said hybridisation pattern to obtain the polynucleotide expression profile of the biological material or the mixture of biological materials. The expression profile in the biological material may be determined to correspond to the expression of specific genes.

The analysis of the expression profile includes several steps of procedures in which well known techniques are used, such as those mentioned in Sambrook et al., Molecular Cloning: A Laboratory approach, Cold Spring Harbour Press, NY (1987), and in Ausubel et al., Current protocols in Molecular Biology, Greene Publishing Co. NY, (1995).

The biological material to be evaluated needs to be identified and isolated such as e.g. described in Example 3. For the ability to perform the analysis cDNA are generally produced from isolated total RNA or polyA RNA (mRNA). The total

RNA/mRNA can be isolated using a variety of techniques. Numerous techniques are well known (see Sambrook et al., Molecular Cloning: A Laboratory approach, Cold Spring Harbour Press, NY (1987), and Ausubel et al., Current protocols in Molecular Biology, Greene Publishing Co. NY, (1995)). In general, these techniques include a first step of lysing the cells and then a second step of enriching for or purifying RNA.

The isolated total RNA/mRNA are reversed transcribed using a RNA-directed

DNA polymerase, such as "reverse transcriptase" isolated from such retroviruses as

AMV, MoMuLV or recombinantly produced. Many commercial sources are available

(e.g., Perkin Elmer, New England Biolabs, Stratagene Cloning Systems). Preferably the mRNA is reversed transcribed into cDNA and at the same time a label is incorporated for later detection of the hybridised amplified products on the array. The amplification by PCR may be performed according to Example 2. The label may vary dependent on the system to be used for the detection and several labels are well known in the area of molecular biology (e.g. radioactive labels, fluorescent labels, coloring labels, chemical labels etc.)

The labelled cDNA is then denaturated and used for hybridisation on the array. The hybridisation conditions vary and are dependent on the aim with the expression profile obtained after the hybridisation. After hybridisation of the labelled cDNA, the specific differential display array is washed to remove the cDNA which have not hybridised to the fragments and the hybridised labelled cDNA are detected by a suitable means and an expression profile obtained.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is further illustrated by reference to the accompanying drawing, in which:

Fig. 1 shows genes selected for optimal differentiation between antidepressant drugs and antipsychotic drugs based on a number of reference compounds. Fig. 2 shows the characterisation of a chemical entity (sertindole) as an antipsychotic drug.

Fig. 3 shows the characterisation of a chemical entity with no therapeutic relevant properties (cocaine).

Fig. 4 shows comparison of the antidepressant drugs and cocaine. Fig. 5 shows genes selected for optimal differentiation between anxiolytic drugs and antipsychotic drugs based on a number of reference compounds.

Fig. 6 shows genes selected for optimal differentiation between anxiolytic drugs and antidepressant drugs based on a number of reference compounds. Fig. 7 shows genes selected for optimal differentiation between anxiolytic drugs and sedative hypnotic drugs based on a number of reference compounds.

EXAMPLES

The invention is further illustrated with reference to the following examples, which are not intended to be in any way limiting to the scope of the invention as claimed.

EXAMPLE 1

Therapeutic potential of a chemical entity as antidepressant or antipsychotic

A test platform for evaluating the therapeutic potential of a chemical entity as antidepressant or antipsychotic may be established as follows: Exemplified outline of the process:

1. A number of array reference compounds are selected from two or more classes of treatments (e.g. antidepressants and antipsychotics).

2. Mice are treated with array reference compounds or control compounds according to standard treatment paradigms (e.g. from the literature).

3. Target tissues (e.g. brains) are isolated from the treated mice and the RNA is extracted using standard RNA extraction techniques.

4. Microarray analysis or differential display (RFDD) analysis is performed for all array reference compounds (compound vs. control), in order to identify genes or gene fragments that are affected by the treatments.

5. A microarray platform is generated using the entire pool of selected genes or gene fragments.

6. This microarray platform is used for generating gene expression profiles of the array reference compounds. 7. THE QUALITY FILTER: The microarray results (profiles) are analysed with GeneSight (BioDiscovery). All microarrays are analysed for signal intensity and signal quality. In this step we isolate the genes that carries reliable information. 8. THE "RIGHT GENES" FILTER: Using mathematical algorithms genes are selected to give the optimal differentiation of the two classes

(antidepressants and antipsychotics) with a predetermined statistical range. Genes are listed according to their contribution in the differentiation of the two classes. In this step we are isolating gene that carries the right information (for optimal classification). 9. The selected genes are spotted on a new array and this is the specific differential display array for depression/schizophrenia.

10. Using the specific differential display array a database is generated based on the gene expression profiles of selected reference compounds from the two classes. The profiles will fall in the two predetermined classes (antidepressants and antipsychotics). This is the screening platform.

11. TESTING THE CHEMICAL ENTITY (compound X): Mice are treated with compound X (or corresponding control) and RNA is isolated from the target tissue.

12. Microarray analysis is performed using the specific differential display array for depression/schizophrenia, and the profile is specific for compound X.

13. Using statistical analysis the profile of compound X will be classified within one of the two classes if it reaches a pre-determined significance level. Otherwise the profile of compound X will remain unclassified.

14. If the profile of compound X is statistically similar to one of the classes (e.g. antidepressant treatments) then compound X has a potential as a new treatment for this disease. 15. In summary, we have evaluated a compound X for potential therapeutic properties using a "drug target independent"- approach.

EXAMPLE 2

Illustration of the invention in differentiation between antidepressant drugs and antipsychotic drugs.

Using a microarray including more than 5000 genes, a number of genes are selected for optimal differentiation between antidepressant drugs (reference compounds: fluoxetine, maprotiline, and imipramine) and antipsychotic drugs (reference compounds: haloperidol, chlorpromazine, and clozapine) (Fig. 1). This is the part of the "database" for characterising the therapeutic potential of a chemical entity. In Fig. 2, a chemical entity is characterized as an antipsychotic drug. In this case, the chemical entity is sertindole, known as an antipsychotic drug.

Fig. 3 shows the characterisation of a chemical entity with no antidepressant or antipsychotic properties (i.e. cocaine). As can be seen, cocaine is relatively more similar to the group of antidepressant drugs than to the group of antipsychotic drugs. To test the resemblance between cocaine and the antidepressant drugs, cocaine was compared with the antidepressant drugs alone: Fig. 4 shows comparison of the antidepressant drugs and cocaine. As can be see, cocaine is not similar to the antidepressant drugs. EXAMPLE 3

Obtaining total RNA from a biological source

Obtaining total RNA from antidepressant treated mice

Male NMRI mice (Age: 12-14 weeks) were treated for 21 days with fluoxetine (10 mg/kg) or clomipramine (15 mg/kg) receiving one intraperitoneal injection per day. Control animals were treated with 0.9% saline in a similar fashion. On day 22, 24 hrs after the last injection the mice were sacrificed by cervical dislocation and the brains were rapidly isolated and snap-frozen in liquid nitrogen. Total RNA was purified from pituitary glands using TRI reagent according to standard procedure (Chomczynski and Sacchi, 1987).

The RNA was extracted using the phenol-chloroform extraction method (Chomczynski and Sacchi, 1987) from brains from three animals. The RNA was evaluated on a gel, and only RNA showing distinct bands on this gel was used for further analysis. Further, the RNA concentration was measured using OD. The RNA was then pooled from two of the animals.

EXAMPLE 4 Identification of differentially regulated gene fragments

Restriction fragment differential display analysis may be carried out as described in patent US 6,261 ,770 (Azign Bioscience A/S). Five hundred (500) ng of total RNA per sample (pooled from 3 animals) is reverse-transcribed using a poly-dT primer, (T25V from DNA Technology, Arhus, Denmark), and 100 units Reverse Transcriptase in 25 μl volume. The reaction is incubated at 42°C for 2 hours. For second-strand synthesis, a 50 μl mix consisting of buffer, dNTPs, 12 units DNA Polymerase I, and 0.8 units RNase H is added. The reaction mixture is incubated at 16°C for 2 hours, and subsequently pheno chloroform extracted, precipitated by ethanol and dissolved in 20 μl H₂0. Ten (10) μl cDNA is checked on an agarose gel for cDNA smear between 100 and 2000 bp.

The remaining 10 μl cDNA is digested with Taql endonuclease at 65°C for 2 hours. The restriction fragment differential display PCR template is completed by ligating the digest with a standard adaptor and an EP-adaptor containing an extension protection group (EPG) at 37°C for 3 hours using T4 DNA ligase. The incubation is carried out at 37°C and not the standard 16°C to maintain the Taql endonuclease active during the ligation. This prevents religation of cDNA Taql-fragments but not adaptor ligation since the latter does not re-establish the Taql site.

The template is PCR amplified using a 0-extension primer complementary to the EP-adaptor in combination with a 3-extension primer recognizing the standard adaptor and the three nucleotides adjacent to the Taql site. The 0-extension primer is kinase- labelled with [³³P]dATP (3000 Ci mmol-1 , equivalent to 1.11 x 1014 Bq mmol-1 , ICN).

All PCR reactions are carried out in 20 μl volume using 0.2 μl template and standard concentration of dNTPs and primers, using the following PCR-amplification profile: precycle 94 °C, 1 min., for the first 10 cycles: 94°C, 30 sec; 60°C with touchdown by

0.5°C for each cycle until 55°C is reached; 72°C, 1 min, for the last 25 cycles: 94°C, 30 sec; 55°C, 30 sec; 72°C, 1 min.

Samples from control and test animals are analysed in parallel and resolved on a standard 6% polyacrylamide sequencing gel by the ALF Express DNA sequencer (Amersham Pharmacia Biotech) and dried down onto Whatman paper. In addition a radio-labelled size marker (Research Genetics) is included on all gels.

Bands corresponding to differentially expressed genes (between samples from control and test animals) are identified and marked on the autoradiogram. The dried gel is lined up with markings on the film and the fragments are excised from the gel. The gene fragments are eluted (in TE buffer), and re-amplified using the same PCR conditions and primers as in the initial PCR reaction.

EXAMPLE 5

PCR fragment cloning, seguencing, and bioinformatic analysis of differentially regulated genes

PCR fragments are cloned using a commercially available cloning kit according to the manufacturers instructions. The selected clones are then sequenced using the Beckman CEQ-2000 sequencer. Sequences are converted to FASTA format and any vector sequence is masked using the 'crossjnatch' program running on a Linux platform. Each sequence is used to search for matches in the GenBank database and the EST subset of Genbank using the BLAST sequence similarity search program. For each sequence the most significant hit is reported as a bit score, length of alignment region, percentage identity, and the GenBank gene name annotation.

EXAMPLE 6

Amplification of the determined fragments

Brain total RNA may be purchased from Clontech (#64040-1 Mouse Brain) The cDNA is synthesized using the Omniscript RT Kit (Qiagen, 205111).

The gene fragments are amplified according to the following procedure:

1-2 μl cDNA obtained above 250 μM dNTP (27-2035-01 , Amersham Pharmacia) 0.5 μM of each PCR primers 1.5 mM MgCI₂ final concentration (Y02016, Gibco BRL) 1X PCR buffer without MgCI₂ (Y02028, Gibco BRL) 2.5 UTaq polymerase (10966-026, Gibco BRL) H₂O to 20 μl final volume

The PCR generated fragments are separated using on a conventional agarose gel and cloned into a suitable vector according to supplier's instruction and the nucleotide sequence is analysed using CEQ 2000 DNA Analysis System (Beckman Coulter, U.S.A.).

EXAMPLE 7

Preparation of Master Glycerol Stocks

The glycerol stocks are prepared in 96 wells-trays (Corning Cat. No. cci3793) on a Biomek. 50μl glycerol media is transferred into each well of a 96-well tray (Corning Cat. No. cci3793). 50 μl bacterial culture is transferred into each well of the plane 96- well tray and mixed with 2 x 100μl. A Storage Mat-I lid (Corning, Cat. No. 3094) is placed on each tray and the trays stored at -80 °C.

EXAMPLE 8

Preparation of plasmids for the specific differential display arrays Ampicillin (100mg/ml) is added to Circlegrow medium (obtained from Bio101 ,

Carlsbad, CA 92008, U.S.A.). Circlegrow medium/ampicillin is added to each well in a 4 x 2 ml 96 deep well tray (Corning Cat.No. cci 3961). The glycerol stocks are added to each well. The tray is sealed with sealing sheet (Merck Eurolab A/S, Denmark), and incubated with shaking at 37°C for 16 hours prior to plasmid purification. The plasmid purification is performed using Biomek or a conventional method according to Maniatis et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbour Press) (1989) well known for a person skilled in the art. Following plasmid purification the specific fragments are generated by standard PCR using specific primer sets. The fragments are made "spot-ready" by precipitation according to standard procedure and dissolving in spot buffer in a suitable concentration.

EXAMPLE 9

Preparation of 3D-Link Amine-Binding slide (array)

3D-Link Amine-Binding slide (array) may be obtained from SurModics, Inc, Minneapolis, USA. To have an orientation on the arrays visible after scanning of the slides each of the arrays corners are marked by a double labelled Cy3 and Cy5 primer with a 5' end amino group (5'-Cy3/5) in a final concentration of 1 pmol/μl to enable the possibility to place a grid on the scanned array. After spotting, seal with sealing tape and store at -20°C until use. 3MM paper is pre-wetted with saturated NaCI solution. Place all the slides in a slide box without a lid. Place the slide boxes in a plastic bag containing the NaCI saturated 3MM paper. Close the plastic bag and incubate the slides. After incubation, remove the slides from the plastic bag. Store at room temperature until blocking of free reactive groups on the slides.

Place the slide in pre-warmed blocking solution (0.1 % SDS, SurModics Blocking Solution. Incubate the slide for 20 minutes at 50 °C. Wash the slide in redistilled H₂O. Incubate the slide in 4 x SSC, 0.1% SDS solution (50°C), and wash the slide in room temperate redistilled H₂O. Incubate the slide boiling redistilled H₂O for 2 min. Wash in redistilled H₂O at room temperature. Incubate the slide in pre-hybridisation buffer prewarmed to 50°C (50 ml 20 x SSC, 10 ml 100 x Denhardt solution, 2 ml 10% SDS, 4 ml Salmon Sperm DNA (10 mg/ml), 134 ml Redistilled H₂O) at 50°C for 30 min. Wash slide in redistilled H₂O. Store the slides at room temperature; dry and dark until use.

EXAMPLE 10

Using the prepared specific differential display arrays for determination of an expression profile in a biological material

Preparation of labelled samples RNA from the biological material is obtained according to the procedures of

Examples 3.

Precipitate the RNA by centrifugation and wash the RNA pellet in 70% ethanol. Precipitate the RNA at 15,000 x g for 15 min. Discard the supernatant and let the pellet air-dry. Adjust the RNA concentration to 1μg/μl with DEPC-H2O. In 2 separate tubes add 25 μl total RNA (1μg/μl) and 7 μl DEPC treated H₂O. Add 4 μl of oligo-dT (e.g. T25V primer) (1 μg/μl) to each tube. Incubate the tubes in a Thermal cycler at 65°C for 3 min.

Prepare tube 1 by adding, 5 μl 10 x cDNA Buffer (500mM Tris-HCI, pH 8.3; 800 mM KCI; 100 mM MgCI2; 40mM DTT), 2 μl Cy3-dUTP (1mM, Cat. No. PA53022, Amersham Pharmabiotech), 5 μl 10 x dNTP (5mM dATP; 5 mM dCTP; 5 mM dGTP; 5mM dTTP). Mix the contents and add 2 μl reverse transcriptase (100 U/μl). Prepare tube 2 by adding, 5 μl 10 x cDNA Buffer (500mM Tris-HCI, pH 8.3; 800 mM KCI; 100 mM MgCI2; 40mM DTT), 2 μl Cy5-dUTP (1mM, Cat. No. PA55022, Amersham Pharmabiotech), 5 μl 10 x dNTP (5mM dATP; 5 mM dCTP; 5 mM dGTP; 5mM dTTP). Mix the contents and add 2 μl reverse transcriptase (100 U/μl). Incubate tube 1 and 2 for 42 °C for 60 min and at 65°C for 15 min. Decrease the temperature to 42°C. Add reverse transcriptase (100 U/μl) to each tube. Incubate at 42°C for 60 minutes and at 65°C for 15 min. Precipitate with 3M Na-acetate and 96% ethanol. Wash each pellet in 80% ethanol. Resuspend each pellet in RNase Mix (10 mM Tris-HCI (pH 7.5), 0.1 mM EDTA (pH 8.0), RNAase A 100 mμ/ml). Incubate at 37°C for 60 min. Add 30 μl sterile H2O to each tube. Precipitate using 3M Na-Acetate (pH 6.0) and ice-cold 96% ethanol. Wash the pellets in 80% ethanol. Resuspend each pellet in 15 μl hybridization buffer (5 x SSC, 0.1 % SDS, 100 μg/ml, blocking RNA). Mix the two fluorescents probes 1:1 in a PCR tube. This is the Sample-Mix.

Hybridisation and detection of hybridisation pattern

Denature the Probe-Mix at 100°C for 3 min. Transfer the Sample-Mix directly to

55°C for 30 sec. Place the Sample-Mix on ice. Add the Sample-Mix to the array slide. Place the slide in a box and inside the petri dish with the pre-wetted 3MM paper.

Replace the lid back onto the petri dish. Place the petri dish in a plastic bag.

Incubate the petri dish in a dark incubator at 65°C 12-16 hours. Wash slides in

Washing Buffer I (2xSSC). Submerge the slides in pre-warmed Washing Buffer II (2 x

SSC, 0.1% SDS. Pre-warm at 65°C for 1 hour in a volume of least 10 ml/slide to cover the slides. Incubate on an orbital shaker at 65°C for 10 min. Wash the slides in

Washing Buffer III (0.2 x SSC) in a volume of least 10 ml/slide to cover the slides.

Incubate on an orbital shaker at room temperature for 3 min. Wash the slides in

Washing Buffer IV (0.1 x SSC) Wash the slides in Washing Buffer V (0.5 x SSC).

Repeat washing with washing buffer V for additional 3 times. Remove all Washing Buffer V by centrifuging at δOOrpm for 3 min. The scanning of the slide and evaluation were performed using Affymetrix 418 Scanner, Affymetrix 418 Scanner Software and the software ImaGene 4.0 (BioDiscovery) according to the user manual.

EXAMPLE 11 Selecting gene fragments according to signal gualitv and information in relation of classification of the reference compounds.

The generated microarrays are analysed using the Genesight Software package (BioDiscovery), and assessed for signal quality and signal intensity. Only those gene fragments that give rise to a signal 100% higher than background level are selected. Using mathematical algorithms genes are selected to give the optimal differentiation of the two classes (antidepressants and antipsychotics) with a predetermined statistical range. Genes are listed according to their contribution in the differentiation of the two classes. In this step we are isolating gene that carries the right information (for optimal classification). A database is generated based on the gene expression profiles of the reference compounds from the two classes. The profiles will fall in the two predetermined classes (antidepressants and antipsychotics). This is the screening platform. EXAMPLE 12

Testing a chemical entity using the micro array platform with selected and gualitv tested gene fragments:

Mice are treated with a chemical entity (compound X) (or corresponding control) and RNA is isolated from the target tissue according to Example 3. Microarray analysis is performed according to Example 10. The expression profile of compound X is compared with the profiles of the reference compounds and evaluated statistically for profile similarity. Compound X will be classified in one of the two classes if it reaches a pre-determined significance level. Otherwise the profile of compound X will remain unclassified. If the profile of compound X is statistically similar to one of the classes (e.g. antidepressant treatments) then compound X has a potential as a new treatment for this disease, and is classified as such.

EXAMPLE 13 Further illustration of the invention in differentiation between anxiolytic drugs, antipsychotic drugs, antidepressant drugs and sedative hypnotic drugs.

Using a microarray including more than 5000 genes, a number of genes are selected for optimal differentiation.

Fig. 5 shows the differentiation between antipsychotic drugs (reference compounds: haloperidol, clozapine, chlorpromazine, flupentixol, and quetiapine) and anxiolytic drugs (reference compounds: lorazepam, alprozalam, flunitrazepam, diazepam, and chlordiaze).

Fig. 6 shows the differentiation between antidepressant drugs (reference compounds: fluoxetine, maprotiline, imipramine, venlafaxin, and citalopram) and anxiolytic drugs (reference compounds: chlordiaze, flunitrazepam, diazepam, lorazepam, and alprozalam).

Fig. 7 shows the differentiation between anxiolytic drugs (reference compounds: lorazepam, flunitrazepam, diazepam, alprozalam, and chlordiaze) and sedative hypnotic drugs (reference compounds: zolpidem and zopiclone.

Claims

1. A method for evaluating a therapeutic potential of a chemical entity comprising:

^■ each reference compound has a known biological effect, and

^■ each distinct class has at least one reference compound as a member;

• generating a gene expression profile of the chemical entity;

2. The method according to claim 1 , wherein the generation of an expression profile of each of the reference compounds comprises:

• obtaining a first expression profile of a biological material not treated with the reference compound;

• generating an expression profile of the reference compound based on differences between the first expression profile and the second expression profile.

3. The method according to claims 1 or 2, wherein the generation of an expression profile of the chemical entity comprises:

4. The method according to any one of claims 1-3, wherein each reference compound may be grouped in one of two distinct classes according to its biological effect.

5. The method according to claim 4, wherein each reference compound may be grouped in one of the two classes of antidepressants and antipsychotics.

6. The method according to claim 4, wherein each reference compound may be grouped in one of the two classes of a therapeutic action with one or more side effects and a therapeutic action without one or more side effects.

7 The method according to claim 6, wherein the therapeutic action is antidepressant and the side effect is nausea, dizziness, anxiety, or drowsiness.

8. The method according to claim 4, wherein each reference compound may be grouped in one of two distinct classes according to its action on a target.

9. The method according to claim 8, wherein each reference compound may be grouped in one of the two classes of 5-HT1 receptor agonists and 5-HT2 receptor agonists.

10. The method according to any one of claims 1-9, wherein the gene expression profiles of the plurality of reference compounds and the gene expression profile of the chemical entity are generated by the use of a differential display technique.

11. The method according to claim 10, wherein the differential display technique used is restriction fragment differential display.

12. The method according to claim 11 , wherein the restriction fragment differential display technique is performed by the use of a specific differential display array comprising: a plurality of polynucleotide spots associated with a surface of a support, wherein the individual polynucleotide spot comprises: an individual gene fragment selected with the purpose to contribute to a optimal differentiation between one of the at least two distinct classes of biological effect.

13. The method according to claim 12, wherein the individual gene fragments are selected by the use of mathematical procedures such as Gene Shaving, Principal

Component Analysis, or statistical analysis.

14. The method according to claims 12 or 13, wherein the individual gene fragments are selected by generating gene expression profiles of a plurality of array reference compounds, wherein:

■ each array reference compound has a known biological effect, and 5 ^■ each array reference compound may be grouped in one of at least two distinct classes according to their biological effect, and

^■ each distinct class has at least one array reference compound as a member.

10 15. The method according to claim 14, wherein the group of array reference compounds and the group of reference compounds have identical members.

16. A specific differential display array for generating a gene expression profile of a test compound comprising: 15 a plurality of polynucleotide spots associated with a surface of a support, wherein the individual polynucleotide spot comprises an individual gene fragment selected with the purpose to contribute to a optimal differentiation between one of at least two distinct classes of biological effect of the test compound.

20 17. The specific differential display array according to claim 16, wherein the individual gene fragments are selected by the use of mathematical procedures such as Gene Shaving, Principal Component Analysis, or statistical analysis.

18. The specific differential display array according to claims 16 or 17, wherein the 25 individual gene fragments are selected by generating gene expression profiles of a plurality of array reference compounds, each array reference compound being grouped in one of the at least two distinct classes according to their biological effect.

19. The specific differential display array according to claim 18, wherein the

30 generation of a gene expression profile of each of the array reference compounds comprises:

• obtaining a first expression profile of a biological source not treated with the array reference compound;

• obtaining a second expression profile of a biological source treated with 35 the array reference compound;

• generating an expression profile of the array reference compound based on differences between the first expression profile and the second expression profile.

20. The specific differential display array according to claim 19, wherein each array reference compound may be grouped in one of two distinct classes according to its biological effect.

5 21. The specific differential display array according to claim 20, wherein each array reference compound may be grouped in one of the two classes of antidepressants and antipsychotics.

22. The specific differential display array according to claim 20, wherein each array 10 reference compound may be grouped in one of the two classes of a therapeutic action with one or more side effects and a therapeutic action without one or more side effects.

23. The specific differential display array according to claim 22, wherein the therapeutic action is antidepressant and the side effect is nausea, dizziness, anxiety,

15 or drowsiness.

24. The specific differential display array according to claim 20, wherein each array reference compound may be grouped in one of two distinct classes according to its action on a target.

20

25. The specific differential display array according to claim 24, wherein each array reference compound may be grouped in one of the two classes of 5-HT1 receptor agonists and 5-HT2 receptor agonists.

25 26. The specific differential display array according to any one of claims 19-25, wherein the gene expression profiles of the plurality of array reference compounds are generated by the use of a differential display technique.

27. The specific differential display array according to any one of claims 16-26, 30 wherein the individual polynucleotide spot is in single stranded or double stranded form.

28. The specific differential display array according to claim 27, wherein the individual polynucleotide spot is of DNA, RNA, cDNA, natural, synthetic, semisynthetic

35 origin or is a chemical analogous such as LNA and PNA.

29. The specific differential display array according to claims 27 or 28, wherein the support is made of a flexible or rigid material.

30. The specific differential display array according to any one of claims 27-29, wherein the array comprises from 4 to 40,000 polynucleotide spots.

31. A database comprising a plurality of gene fragments being present on the specific differential display array according to any one of claims 16-26.

32. A chemical entity evaluated by the method according to claims 1-15.