WO2007023903A1

WO2007023903A1 - Method for predicting carcinogenicity of test substance

Info

Publication number: WO2007023903A1
Application number: PCT/JP2006/316618
Authority: WO
Inventors: Hiroshi Matsumoto; Yoshikuni Yakabe; Yoshihisa Sudo; Koichi Saito; Kayo Sumida; Koji Nakayama; Masaru Sekijima; Tomoyuki Shirai
Original assignee: Chemicals Evaluation And Research Institute; Sumitomo Chemical Co., Ltd.; Mitsubishi Chemical Safety Institute Ltd.
Priority date: 2005-08-26
Filing date: 2006-08-24
Publication date: 2007-03-01
Also published as: JP2007054022A; JP4806234B2

Abstract

A method for predicting carcinogenicity of a test substance comprising a first step in which a plurality of carcinogens is administered to each group administered with each carcinogen and mRNA is collected from each group administered with each carcinogen after a lapse of a predetermined period of time and expression altered genes whose mRNA expression level is increased or decreased are selected, a second step in which the expression patterns of the expression altered genes are compared among each group administered with each carcinogen and the expression patterns are classified into a plurality of groups according to the similarity of the expression patterns and the expression patterns of the expression altered genes are prepared for each group and a third step in which a test substance is administered to a group administered with a test substance and mRNA is collected from the group administered with a test substance after a lapse of a predetermined period of time and expression patterns are obtained for the expression altered genes and compared with the expression patterns of the expression altered genes prepared in advance for each group of the groups administered with carcinogens and a concordance degree thereof is calculated and the carcinogenicity of the test substance is predicted.

Description

Specification

Method for predicting carcinogenicity of test substance

Technical field

The present invention relates to a carcinogenicity prediction method for administering a test substance to an animal and predicting the carcinogenicity of the test substance in terms of the amount of mRNA expressed in the test substance administration group.

Background art

[0002] In order to evaluate the toxicity of chemical substances! Long-term toxicity such as carcinogenicity requires animal experiments that require a large amount of cost and a long test period. In addition, when extrapolating the results of animal experiments to humans, there are various problems such as drought that must take into account species differences and mechanisms of action.

[0003] On the other hand, due to the remarkable development of technologies related to genomic information in recent years, the hazard assessment of chemical substances has been carried out at the gene level. For example, Patent Document 1 describes a method for predicting a toxic effect of a chemical substance by detecting a difference in gene expression between tissues and cells exposed to the chemical substance.

[0004] Since the presence of genes involved in carcinogenic mechanisms is also expected in the evaluation of carcinogenicity of chemical substances, it is possible to evaluate at the gene level by detecting differences in the expression of these genes. It is believed that there is. However, at the present stage, the carcinogenic mechanisms at the gene level caused by chemical substances are hardly elucidated, and it is very difficult to predict the carcinogenicity of chemical substances.

[0005] Recently, a method using a DNA microarray has been developed as a method for comprehensively collecting and analyzing gene expression profile data, and has been used in various fields. A DNA microarray (DNA chip) is composed of a substrate such as a glass slide and hundreds of thousands of kinds of DNA immobilized on the substrate. Obtain information on the expression level of mRNA by binding cDNA or cDNA prepared from mRNA using the complementarity of nucleic acid molecules to DNA immobilized on the DNA microarray substrate. Is possible.

[0006] DNA microarrays equipped with known genes have already been commercialized and generally used. Is in an environment where However, commercially available arrays are not designed or built for the purpose of chemical toxicity assessment. The use of DNA microarrays at normal toxicology scales requires significant costs to build the entire system, including data analysis. For these reasons, DNA microarrays are not widely used in the toxicity test field.

Patent Document 1: Japanese Patent Application Laid-Open No. 2003-304888 (Claims)

Disclosure of the invention

Problems to be solved by the invention

[0007] In general, in an animal experiment for evaluating the carcinogenicity of a chemical substance, the chemical substance is continuously administered until cancer appears in the test animal or the test animal dies. Since cancer appears after a long incubation period, long-term animal experiments are required.

However, even before the appearance of cancer, it is expected that some change occurs at the gene level by being stimulated by a carcinogen.

[0009] The object of the present invention is to conduct long-term animal experiments by detecting differential expression and analyzing data for genes that are likely to be involved in the carcinogenic mechanism in the early stages of cancer development. It is intended to provide a method for predicting the carcinogenicity of a test substance without any problems.

Means for solving the problem

[0010] The present inventors paid attention to mRNA whose expression level fluctuates when a carcinogen is administered to animals, and tried to predict the carcinogenicity of the test substance. Initially, we thought that these mRNA increase and decrease patterns were similar among carcinogens. However, in reality it behaved differently and it was hard to share one similarity.

[0011] Therefore, the present inventors grouped carcinogens with similar mRNA increase / decrease patterns (that is, gene expression patterns) to obtain an expression pattern for each group in advance, and then added a test substance. An attempt was made to compare the expression pattern when administered with the expression pattern of each group of carcinogens and evaluate the degree of agreement. Then, it was found that the carcinogenicity of the test substance can be predicted with high accuracy even in the initial stage of animal experiments, and the present invention has been completed. [0012] That is, the present invention for solving the above problems is described below.

[0013] [1] A plurality of carcinogens are administered to each carcinogen-administered group, and after a predetermined period of time, each carcinogen-administered group force mRNA is collected, and the mRNA expression level is measured to determine the mRNA expression level. A first step of selecting an expression variable gene with significantly increased or decreased; and

The expression pattern of the expression variation gene selected in the first step is compared among the carcinogen-administered groups, and the expression pattern is classified into multiple groups based on the similarity of the expression pattern, and the expression variation is classified into multiple groups. A second step of preparing a gene expression pattern;

A test substance is administered to a test substance administration group, and mR from the test substance administration group after a lapse of a predetermined period

Change in expression of carcinogen-administered group in which NA was collected and expression pattern of expression-variable gene was obtained, and expression pattern of expression-variable gene of test substance administration group was prepared in the second step in advance A third step of calculating the degree of coincidence compared to the gene expression pattern;

Method for predicting carcinogenicity of test substance having

[0014] [2] A plurality of carcinogens are administered to each carcinogen-administered group, and after a predetermined period, each carcinogen-administered group force mRNA is collected, and the mRNA expression level is measured to determine the mRNA expression level. A first step of selecting an expression variable gene with significantly increased or decreased; and

A test substance is administered to a test substance administration group, and mR from the test substance administration group after a lapse of a predetermined period.

Change in expression of carcinogen-administered group in which NA was collected and expression pattern of expression-variable gene was obtained, and expression pattern of expression-variable gene of test substance administration group was prepared in the second step in advance A third step of calculating the degree of coincidence in comparison with the gene expression pattern; A method for predicting the carcinogenicity of a test substance having the following three steps: Carcinogen administration group strength Fluorescently labeled cDNA or cRNA prepared using the collected mRNA is hybridized to a DNA microarray. And the fluorescence pattern obtained by hybridizing fluorescence-labeled cDNA or cRNA prepared using the collected mRNA to a DNA microarray. A step of comparing and selecting an expression variation gene by detecting a significant difference from the fluorescence intensity of each spot of the fluorescence pattern,

Multiple carcinogen-administered group strength Each of the carcinogen-administered groups was compared by comparing the fluorescent patterns obtained by hybridizing fluorescence-labeled cDNA or cRNA prepared using the collected mRNA to a DNA microarray. Expression between carcinogen-administered groups, comparing expression patterns of variable genes and classifying the expression patterns of carcinogen-administered groups into multiple groups,

Fluorescence pattern obtained by hybridizing fluorescence-labeled cDNA or cRNA prepared from mRNA collected from carcinogen-administered group to DNA microarray, and mRNA force collected from test substance-administered group Prepared fluorescence Comparison of the expression variation gene of the carcinogen-administered group and the expression-variable gene of the test substance-administered group by comparison with the fluorescence pattern obtained by hybridization of labeled cDNA or cRNA to the DNA microarray A method for predicting the carcinogenicity of a test substance, comprising the step of comparing the expression patterns of and calculating the degree of coincidence thereof.

[3] DNA microarray power Test animal power administered with a carcinogen Extracting mRNA, and amplifying or amplifying all or part of the cDNA sequence synthesized by reverse transcription from mRNA with increased or decreased expression level [2] The method for predicting the carcinogenicity of a test substance according to [2], which is a DNA microarray equipped with chemically synthesized DNA.

[4] Any one of [1] to [3], wherein the expression variable gene is selected from five or more genes having a base sequence complementary to the base sequence described in SEQ ID NOs: 1 to 2844 The method for predicting the carcinogenicity of a test substance according to claim 1.

[0017] [5] Carcinogen power to be administered to carcinogen administered groups Clofibrate, di (2-ethylhexyl) phthalate, carbon tetrachloride, 2,4-diaminotoluene, quinoline, phenobarbiter , Jetylnitrosamine, 2-nitropropane, N-nitrosomorpholine, aldrin, di (2-ethylhexyl) adipate, ethur estradiol, hexacyclobenzene, _a- hexachlorocyclohexane, trichloroethylene, Butylated hydroxy vinylol, limonene, safrole, 1,4-dichlorobenzene, 1,4-dioxane, furan, methyl carbamate, thioacetamide, N-nitrosodimethylamine, N-nitrosopiperidine, 2-amino- 3,8-dimethyl imimidazo [4,5-f] quinoxaline, 2-amino-1-methyl-6-fe-louimidazo [4,5-b] -pyridine, benz (a) anthracene, 7, 12-dimethylbenz Anthracene, 3-methylcholanthrene, 4-nitroquinoline-1-oxide, N-ethyl-N-nitrosourea, trichloroacetic acid, tannic acid, urethane, pen Chlorethane, black mouth form, benzo (a) pyrene, N-methyl-Ν'-nitro-N-nitrosoguanidine, tetrachloroethylene, acetamide, jetylstilbestrol, pheotoin sodium salt, D, L-ethionine, 4 -The method for predicting the carcinogenicity of a test substance according to any one of [1] to [4], comprising at least four selected from dimethylaminoazobenzene and chlorinated acid power.

[0018] [6] The test substance according to any one of [1] to [4], wherein the carcinogen administered to the carcinogen administered group includes at least six carcinogens according to [5]. Carcinogenicity prediction method.

[0019] [7] The test substance according to any one of [1] to [4], wherein the carcinogen administered to the carcinogen administration group includes at least nine carcinogens according to [5]. Carcinogenicity prediction method.

[0020] [8] Non-carcinogenic substance power administered to non-carcinogen-administered group 2,6-Diaminotoluene, 8-Hydroxyquinoline, D-mannthol, L-ascorbic acid, 2-clo oral ethanol, 2- (Chloromethyl) pyridine hydrochloride, cQ-menthol, 4-nitro-0-phenol-diamine, benzoin, odoform, lithocholic acid, lindane, 2-chloro-p-phenol-diamine amine sulfate, p-phenylene Diamine dihydrochloride, 2,5-toluenediamine sulfate, aspirin, 4- (chloroacetyl) acetaldehyde, phthalamide, force prolatatam, 1-chloro-2-2-propanol, 3-chloro-p-toluidine The method for predicting carcinogenicity of a test substance according to any one of [2] to [7], comprising at least one selected from the group consisting of glutaraldehyde, 4-nitroanthralic acid, and 1-tronaphthalene.

[0021] [9] The test according to any one of [2] to [7], wherein the non-carcinogenic substance administered to the non-carcinogen-administered group includes at least five non-carcinogenic substances according to [8]. Carcinogenicity prediction of substances Method.

[0022] [10] The test substance according to any one of [2] to [7], wherein the non-carcinogen administered to the non-carcinogen-administered group contains at least 10 non-carcinogens according to [8]. Carcinogenicity prediction method.

[0023] [11] Carcinogenicity prediction of the test substance according to any one of [1] to [10], wherein the expression pattern of the expression variation gene in the carcinogen-administered group is classified by cluster analysis or a decision tree. Method.

[12] The expression pattern of the expression variable gene is classified into three or more groups [1] to [1]

[11] The method for predicting carcinogenicity of a test substance according to any one of [11].

[13] The carcinogen is administered to each carcinogen-administered group, and after 1 to 90 days have passed, each carcinogen-administered group force mRNA is collected. [1] to [12] A method for predicting the carcinogenicity of a test substance.

[0026] [14] After administering a carcinogen to each carcinogen-administered group, collect the mRNA of each carcinogen-administered group after 14 to 28 days, [1] to [12] A method for predicting the carcinogenicity of a test substance.

[15] The test substance-administered group force is also sampled after 1 to 90 days after the test substance is administered to the test substance-administered group, [1] to [14] A method for predicting the carcinogenicity of a test substance.

[0028] [16] The test substance administration group force mRNA is collected after 14 to 28 days after administration of the test substance to the test substance administration group [1] to [14] A method for predicting the carcinogenicity of a test substance.

The invention's effect

[0029] According to the present invention, the carcinogenicity of the test substance is predicted by comparing the mRNA expression pattern of the test substance administration group with the mRNA expression pattern of each group of the carcinogen administration group. can do. In the prediction method of the present invention, since the carcinogenicity of the test substance is predicted from the expression level of mRNA, it is not necessary to conduct a long-term animal experiment in which the compound is continuously administered until cancer appears in the test animal. Carcinogenicity of chemical substances can be easily predicted by animal experiments in a short period of about 1 to 90 days. Brief Description of Drawings

[0030] FIG. 1 is a part of a saddle diagram in which chemical substances are classified by cluster analysis in Example 1.

FIG. 2 is another part of the saddle diagram shown in FIG.

FIG. 3 is a part of a saddle diagram in which chemical substances are classified by cluster analysis in Example 1.

4 is another part of the saddle diagram shown in FIG.

FIG. 5 is a part of a saddle diagram in which chemical substances are classified by cluster analysis in Example 1.

FIG. 6 is another part of the saddle diagram shown in FIG.

FIG. 7 is a saddle diagram in which chemical substances are classified by cluster analysis in Example 1.

FIG. 8 is a saddle diagram in which chemical substances are classified by cluster analysis in Example 1.

FIG. 9 is a saddle diagram in which chemical substances are classified by cluster analysis in Example 1.

FIG. 10 is a flowchart showing the carcinogenicity prediction method constructed in Example 1.

FIG. 11 is a saddle diagram in which chemical substances are classified by a decision tree in Example 2.

BEST MODE FOR CARRYING OUT THE INVENTION

[0031] The method for predicting carcinogenicity of a test substance of the present invention is performed according to the following procedure.

[0032] First, after administering a plurality of carcinogens to each carcinogen-administered group, tissue of the test animal is collected for each group, and the expression level of each mRNA is measured.

[0033] Although the number of carcinogens administered to the carcinogen administered group is 4 or more, it is preferable to use 6 or more, and more preferably 9 or more to improve the accuracy of prediction. There is no upper limit on the number of carcinogens because the greater the number of carcinogens used, the higher the accuracy of predicting the carcinogenicity of the test substance, but the carcinogenicity of the test substance is sufficiently accurate if about 30 substances are administered. Can be predicted.

[0034] As the carcinogen administered to the carcinogen-administered group, a known chemical substance that has been confirmed to be carcinogenic can be administered. For example, clofibrate, di (2-ethylhexyl) phthalate, carbon tetrachloride, 2,4-diaminotoluene, quinoline, phenobarbital, dimethyl ditrosamine, 2-nitropropane, N-nitrosomorpholine, aldrin, Di (2-ethylhexyl) adipate, ethynyl estradiol, hexane cyclobenzene, α-hexachlorocyclohexane, trichloroethylene, butylated hydroxy-sol, limonene, safrole, 1,4-dichlorobenzene, 1 , 4-dioxane, furan, methyl carbamate, thio Oacetamide, N-nitrosodimethylamine, N-nitrosopiperidine, 2-amino-3,8-dimethylimidazo [4,5-f] quinoxaline, 2-amino-1-methyl-6-phei-louimidazo [4,5 -b] -pyridine, benzene (a) anthracene, 7, 12-dimethylbenzanthracene, 3-methylcholanthrene, 4-nitroquinoline-1-oxide, N-ethyl-N-nitrosourea, triclo oral acetic acid, Tannic acid, urea, pentachloroethane, black mouth form, benzo (a) pyrene, N-methyl-Ν'-nitro-N-nitrate mouth soguanidine, tetrachloroethylene, acetamide, jetylstilbestrol, fe

-Toin sodium salt, D, L-ethionine, 4-dimethylaminoazobenzene, and chlorendic acid.

[0035] When a carcinogen is administered to a group of animals, a carcinogen solution in which the carcinogen is dissolved in a medium is prepared.

[0036] The medium for dissolving the carcinogen can be used without particular limitation as long as it is a non-carcinogenic medium in which the carcinogen is dissolved. For example, corn oil, purified water, and the like can be used.

[0037] The concentration of the carcinogen solution is preferably set to a concentration that causes an appropriate increase or decrease in the expression level of mRNA that varies characteristically by stimulation of the carcinogen.

[0038] The carcinogen solution is administered to each group continuously by repeating it for a predetermined period of about 1 to 90 days, 1 to several times a day (preferably once a day).

[0039] The administration method of the carcinogen solution to the carcinogen administration group is not particularly limited, and general-purpose methods such as oral administration, intraperitoneal administration, and intravenous administration can be used.

[0040] Rats, mice, dogs, monkeys, guinea pigs, rabbits, etc. can be used as test animals.

[0041] The period from the administration of the carcinogenic substance to the collection of the test animal tissues in each group is about 1 to 90 days. It is preferable to set it to about 13 to 14 days in the medium term. Liver, intestine, lung, kidney, stomach, spleen, brain, blood and the like can be used as test animal tissues to be collected in order to measure the expression level of mRNA.

[0042] After extracting and purifying mRNA by a known method such as tissue strength of test animals, the expression level of each mRNA is measured. Methods for measuring the expression level of mRNA include DNA microarrays and microplates in which cDNA or DNA having a sequence complementary to mRNA is immobilized. Known methods such as a method of binding a fluorescently labeled cDNA or cRNA prepared by mRNA to a yeast, Northern blotting, quantitative RT-PCR, RNase protection assay and the like can be used. Above all, the method using DNA microarray can simultaneously acquire the expression level data for many mRNAs in each carcinogen-administered group, and it is easy to compare the expression patterns between groups. It is.

[0043] For the measurement of the expression level of mRNA, a commercially available microarray selected according to the type of test animal can also be used. For example, when a rat is used as a test animal, Gene Chip (trade name, manufactured by Affymetritas), Rat Oligo Microarray Kit (trade name, manufactured by Agilent) and the like can be used.

[0044] A method for grouping gene expression patterns from mRNA expression levels will be described below. A gene expression pattern is a collection of data composed of gene types and their expression levels.

[0045] First, for each carcinogen administered group, a gene whose expression level is significantly increased or decreased (varied) by administration of the carcinogen (hereinafter referred to as an expression variable gene) is selected. The determination of whether the gene expression level changed significantly in each group was based on the relative expression of the non-carcinogenic substance administered group dissolved in the medium used to prepare the carcinogen solution. Perform by comparison. Non-carcinogenic substances include NTP (National Toxicology Program; EPA

Among the compounds officially evaluated as non-carcinogens by IAR (International Agency for Research on Cancer), etc., at least 1 or more, preferably 5 or more, more preferably 10 or more compounds are used. There is no upper limit to the number of non-carcinogenic substances to be used, but sufficient accuracy can be achieved by administering about 20 substances. Non-carcinogenic substances include, for example, 2,6-diaminotoluene, 8-hydroxyquinoline, D-mannthol, L-ascorbic acid, 2-chloromethyl ethanol, 2- (chloromethyl) pyridine hydrochloride, cQ-menthol, 4 -Nitro-0-phen-diamine, benzoin, odoform, lithocholic acid, lindane, 2-chloro-P-phen-diamine amine sulfate, P-phenol-diamine dihydrochloride, 2,5-toluene diamine sulfate, Aspirin, 4- (chloroacetyl) acetolide, phthalamide, force prolatatam, 1-chloro-2-propanol, 3-chloro-p-toluidine, glutaraldehyde, 4-nitroanthracic acid, and 1- Tronaphthalene etc. can be mentioned [0046] For the analysis of expression patterns, all expression level data of expression variable genes whose expression levels fluctuated in each carcinogen-administered group may be used! /. Since there are many genes whose expression levels fluctuated and the analysis operation becomes complicated, it is necessary to select genes whose expression levels vary greatly between the carcinogen-administered group and the non-carcinogen-administered group. It is preferable to select and use.

[0047] The expression variable gene used for classification of the expression pattern is more preferably at least about 10 to 150, and particularly preferably about 30 to 120 U.

[0048] When rats are used as test animals, SEQ ID NOs: 1 to 2844 show base sequences showing a part of the base sequences of genes whose expression levels tend to vary in the carcinogen-administered group. When the expression variable gene is selected so as to contain 5 or more, preferably 7 or more, more preferably 10 or more of these genes, the accuracy of predicting the carcinogenicity of the test substance can be increased.

[0049] With respect to the selected expression variation gene, the expression pattern is compared between the carcinogen-administered groups, and the expression pattern is classified by a statistical method according to the similarity.

[0050] Examples of statistical methods for classifying expression patterns include analysis methods such as cluster analysis and decision trees.

[0051] When cluster analysis is used, expression patterns can be classified by the following method, for example.

[0052] First, cluster analysis is performed on the expression pattern of the selected expression variable gene. After that, change the selection condition of the expression variation gene, select a different expression variation gene, and perform the cluster analysis even if the combination of other expression variation genes. Select carcinogen groups that are classified into the same cluster even if the combination of expression variation genes is changed. It can be presumed that the carcinogen group that stably forms the same cluster even if the combination of expression variable genes is changed has the same carcinogenic mechanism that does not occur in the cluster formed by chance by the selected expression variable gene. In addition, a group is formed! /, !!, and the remaining carcinogens are newly selected. Expression variation genes are newly selected, and cluster analysis and grouping are similarly performed for the remaining carcinogens. Do the trap. Stable operation of this operation Repeat until the group of carcinogens that forms the star is no longer detected, and use the expression pattern of the expression variable gene that is common to the group of carcinogens as the expression pattern of the group.

[0053] The expression pattern of the carcinogen is classified, and the expression pattern of the expression variable gene is obtained for each classified group. When predicting the carcinogenicity of a test substance, the test substance is administered to the test substance administration group under the same conditions as the carcinogen, and the expression level of the expression variable gene in the test substance administration group is measured after a predetermined period of time. To do.

[0054] The expression pattern of the expression variable gene obtained by administering the test substance is compared with the expression pattern of each expression variable gene obtained by administering the carcinogen, and the expression of each group is compared. Evaluate the degree of coincidence with the pattern. If the expression pattern of the test substance is high / matched with the expression pattern of the group of carcinogens of 1%, the test substance is judged to have carcinogenicity and the degree of match with any expression pattern is low. In some cases, it is determined that there is no carcinogenicity.

[0055] As a method for evaluating the degree of coincidence of expression patterns, for example, a method of evaluating from the hierarchy of the morphological map obtained by performing cluster analysis or decision tree analysis together with carcinogens, expression of expression variable genes For example, a method for calculating a correlation between quantities can be used.

[0056] Hereinafter, an example of a method for predicting the carcinogenicity of a test substance using a DNA microarray will be specifically described.

[0057] First, in order to obtain a gene to be mounted on a DNA microarray, cDNA having a sequence complementary to mRNA is synthesized by reverse transcription reaction from mRNA extracted from the tissue strength of a test animal administered with a carcinogen. When synthesizing cDNA from mRNA, reverse transcription PCR method (RT—PCR), which uses PCR primers obtained from sequence information to amplify each mRNA in a cage shape, reverse transcriptase, etc. Can be used.

[0058] Next, cDNA synthesized using a DNA microarray preparation device was coated on a slide glass that had been surface-treated with polycations such as poly-L-lysine and polyalkylamine in advance, and the DNA microarray was coated. Is made.

[0059] Tissue strength of carcinogen-administered group and tissue strength of medium control group Extracted mRNA and fluorescently labeled cDNA or cRNA obtained with different fluorescent substances were obtained. Hybridize them to a DNA microarray.

[0060] Examples of fluorescent substances used for mRNA labeling include Cy3 (red) and Cy5 (green). As a labeling method, there are known methods such as a method using a primer having a fluorescent material introduced at the 5 ′ end, a method of incorporating a deoxynucleotide derivative of the fluorescent material with a reverse transcriptase, and a method using a post-labeling reagent. Can be adopted.

[0061] A fluorescently labeled cDNA or cRNA hybridized DNA microarray is read with a scanner, and the fluorescence intensity of each spot is measured.

[0062] Even in the non-carcinogen-administered group, the fluorescence pattern of the DNA microarray was obtained in the same manner as in the carcinogen-administered group, and the fluorescence wavelength and fluorescence intensity power of each spot obtained were also expressed. To do.

[0063] Since the mRNA of the carcinogen-administered group and the mRNA of the vehicle control group are labeled with different fluorescent substances, the comparison of the fluorescence wavelength and fluorescence intensity of each spot read by the scanner shows that the mRNA of each group The expression level can be easily compared.

[0064] In the next step, using a statistical method such as cluster analysis or decision tree, the fluorescence pattern of the selected expression variation gene is classified, and each group is compared with the fluorescence pattern of the test substance administration group. The carcinogenicity of the test substance is predicted by evaluating the degree of coincidence.

[0065] The DNA mounted on the DNA microarray includes, in addition to DNA obtained by amplifying the whole or part of the cDNA sequence synthesized by reverse transcription from mRNA extracted from the tissue of the test animal by PCR or the like, DNA obtained by chemically synthesizing part of the cDNA sequence can also be used. The sequence length of the DNA mounted on the DNA microarray is about 20 to 500 bp. 1S is preferably 50 to 400 bp, more preferably 60 to 300 bp.

Example

[0066] Production Example l (Preparation of cDNA microarray)

(1) Preparation of full-length cDNA clones and libraries (RIKEN method) Carcinogen-treated adult animals (6 weeks old) Danaform Inc. was prepared from liver, kidney, spleen and colon tissue from rats A full-length cDNA clone was obtained using a rat full-length cDNA library. The outline of the procedure for preparing the rat full-length cDNA library was as follows. [0067] Four known carcinogens (Diethylnitros oamine: dose 100 mg / kg / day: administration period 3 hours, 1 day, 3 days, 2-amino-3) , 8-dimethyl imadazo [4,5- flquinoxaline: Dose 20 mg / kg / day: Dose duration 3 hours, 1 day, 4 days, Clofi brate: Dose 100 mg / kg / day: Dose duration 1 day, 20, 4 days, Phenobarbital: dose 80 mg / kg / day: administration period 1 day, 2 days, 4 days), and liver, kidney, spleen and large intestine were collected. Total RNA was extracted from the collected tissues by the AGPC method, and mRNA was purified using MACS mRNA Isolation Kit (Miltenyi Biotec). Thereafter, reverse transcription reaction was performed with reverse transcriptase that had been heat-resistant in the presence of trehalose using Anchored oligo (dT) as a primer. This Anchored oligo (dT) primer was tagged with information about the origin of the library and cDNA using a 6-base tag sequence. Next, a full-length cDNA was selected using a cap trapper method, followed by second strand synthesis and restriction enzyme cleavage, followed by orientation and cloning to create a full-length cDNA library. This full-length cDNA library was subtracted by using a cDNA library derived from a compound-untreated rat organ as a driver to increase its specificity.

[0068] (2) Selection of genes to be loaded on cDNA microarray

The homology between clones was examined from the 3'-end nucleotide sequence of a clone in a full-length cDNA library derived from a carcinogen-treated adult animal. The total number of genes derived from carcinogen-treated adult animals was 15,762, of which 4,139 were derived from the liver. The number of non-overlapping base sequences based on UniGene ID was 6,954.

[0069] 8,862 clones were selected from the following criteria as gene pools for selection of genes carrying clones derived from untreated rat organs and clones derived from carcinogen-treated rat organs.

[0070] a) A gene whose expression level fluctuates by 2 times or more or 1/2 of that of a control animal by administration of a carcinogen.

[0071] b) Carcinogen-treated rat organ-derived clone.

[0072] c) Genes reported to be related to carcinogenicity in literature. These genes include oncogenes, tumor suppressor genes, transcription factors, and growth factors. It is a gene that has been reported to be related to carcinogenicity in the literature. For 24 genes that are not in the offspring pool but are considered important for carcinogenicity assessment, clones were purchased or isolated individually.

[0073] d) A gene commonly mounted on a commercially available microarray for toxicity evaluation. As sales arrays, we investigated Clontech Rat Toxicology.2 array, TAKARA Rat Toxicology CHIP ver.O, and NIEHS Rat Chip-MWG Rat Liver-Mergen Rat R01.

[0074] e) Three genes (beta-actin, GAPDH, ubiquitin) as a positive control and lambda phage gene as a negative control.

[0075] The genes included in a) were 2,890 clones, b) included 5,759 clones, c) included 44 genes, and d) included 185 clones.

[0076] (3) Preparation of cDNA microarray

Partial sequences that are non-complementary to each other were designed from the base sequences of the selected genes, and the gene fragments were amplified by the polymerase chain reaction (PCR) method using the 8,862 clones selected in (2) as templates. The length of the amplified gene was designed to be about 300 bp, excluding some genes. Primers for amplifying gene fragments to be loaded on cDNA microarrays by PCR were designed using the Takara Bio Inc. design algorithm. PCR reaction was performed using 35 pmol of each of the forward primer and reverse primer, 2.5 units of Ex-Taq polymerase and a buffer attached to the enzyme, and a reaction volume of 100 L. The reaction was carried out for 37 cycles of 95 ° C for 45 seconds, 95 ° C for 45 seconds, 54 ° C for 30 seconds, 72 ° C for 60 seconds, and further for 72 ° C for 3 minutes. After completion of the reaction, a PCR product was confirmed by subjecting a part of the reaction solution to agarose electrophoresis.

[0077] After the PCR reaction, 8,261 clones were amplified. As a result of detailed examination of the products, the final analysis target was 8,051 clones, except for PCR products that were not single gene fragments or smears. As a result of re-ananotation using the 2003/01/06 version of UniGene database, the number of non-overlapping UniGene IDs was 6,353.

[0078] The obtained PCR product was purified with QIAGEN QIAquick PCR Purification Kit and then spotted on a glass substrate to obtain a cDNA microarray. Two arrays per glass were installed. [0079] Example 1 (carcinogenicity prediction by cluster analysis)

(1) Acquisition of total RNA

Each chemical substance shown in Tables 1 to 4 was dissolved in a medium to prepare a solution. Tables 1 to 4 show the chemical substances, their substance numbers, carcinogenicity, used media, and doses administered to test animals.

[0080] [Table 1]

table 1

Table 2

Table 3

Table 4

Five-week-old male rats (F344, SPF strain) obtained from Nippon Chirus Riva Co., Ltd. were divided into 4 groups Z, and the prepared solution or vehicle was orally administered to the rats of each group. Administration was once a day for 1, 3, 7, 14, 28 days. The liver of each individual was collected on days 2, 4, 8, 15, or 29 of the administration platform. A fragment was excised from the liver and immersed in RNAlater (registered trademark) RNA Stabilization Reagent, a reagent for stabilizing and storing total RNA. At 24 hours at room temperature And then stored at -20 ° C while immersed in RNAlater RNA Stabilization Reagent

[0085] Total RNA was extracted from rat tissue and purified using QIAGEN BioRobot 3000 using QIAGEN RNeasy Midi Kit. The purification method was performed according to the manufacturer's method of use. The outline is as follows.

[0086] 50-100 mg of the excised tissue sample and crushing beads (diameter 5 mm, manufactured by Zircoa) were placed in a 2 mL Eppendorf tube. After adding RLT buffer, tissue disruption was performed twice at 25 Hz for 4 minutes using a Mixer Mill (QI AGEN). Centrifugation was performed at 5,000 × g for 5 minutes to precipitate uncrushed material. 70% ethanol was added to the supernatant, mixed, and added to RNeasy Midi column. After centrifugation at 3,000-5,000 X g for 5 minutes, RW1 buffer was added. After centrifuging at 3,000-5,000 X g for 5 minutes, DNase I solution (QIAGEN) was added and left at room temperature for 15 minutes to add RW1 knocker. After centrifugation at 3,000-5,000 × g for 5 minutes, RPE buffer was added. After centrifugation at 3,000-5,000 X g for 2 minutes, RPE buffer was added again. After centrifuging at 3,000-5,000 X g for 2 minutes, centrifuge at 3,000-5,000 X g for an additional 5 minutes to dry the column. After adding RNase-free water and allowing to stand for 5 minutes, the RNA solution was eluted by centrifugation at 3,000-5,000 X g for 2 minutes. In order to improve the concentration and recovery rate, the eluted RNA solution was again added to the column and allowed to stand for 5 minutes, and then centrifuged at 3,000-5,000 X g for 2 minutes to obtain the final purified RNA solution.

[0087] The 260 nm / 280 nm ratio indicating the purity of the purified total RNA was measured using an optical measuring device. The purified total RNA was examined for its electrophoretic pattern using Agilent BioAnalyser2100. For the electrophoresis, an RNA6000nano chip (Agilent) was used, and the operation was performed according to the method used by the manufacturer. The concentration of purified total RNA was measured by a method using an optical measuring device.

[0088] When the optical path length was 1 cm as an optical measuring apparatus, the calculation was made as 1 {absorbance (260 nm) -absorbance (320 nm)} = 40 ^ g RNA / mL.

[0089] (2) Preparation of fluorescently labeled cDNA

Using 20 or 10 μg of purified total RNA, fluorescently labeled cDNA was prepared using CyScript reverse transcriptase. [0090] Fluorescent labeling was performed using total RNA or mRNA (self-purified, manufactured by Clontech or Kitayama Labes Co., Ltd.) under the following reaction conditions. Final reaction volume was 50 L

[0091] The final concentration of the reaction solution is 10-20 μg of total RNA, Anchored Oligo (dT) (manufactured by Amersham Bioscience) 0.075 μgo dCTP Nucleotide Mix 1 L, 1 X CyScript buffer (manufactured by Amersham Bioscience) 10 mM DTT ゝ Cy3- dCTP or Cy5-dCTP (Amersham Bioscience) 1 nmol.

[0092] A mixture of RNA and Anchored Oligo (dT) was allowed to stand at 70 ° C for 5 minutes, and then allowed to stand on ice for 1 minute. After adding other components, CyScript reverse transcriptase (manufactured by Amersham Bioscience) was added with 100 units and allowed to stand at 42 ° C for 90 minutes in the dark. 12.5 μL of 1 N NaOH was added, and the mixture was left to stand at 65 ° C for 10 minutes to digest the RNA. After 1 N HC1 15 L was added and neutralized, the fluorescent label was purified using the MinElute PCR Purification Kit manufactured by QIAGEN. In order to remove unreacted fluorescently labeled nucleic acid, washing with a PE nozzle was performed twice.

[0093] The administration group (high dose, low dose) was labeled with Cy3, and the control group was labeled with Cy5. Cy5-labeled cDNA, which was also produced in the control group of 4 animals, was mixed in equal amounts at the end of purification, and the mixed-solution power was also dispensed to prepare the hybridization solution.

[0094] (3) Hybridization to cDNA microarray and detection

After mixing to 12 μL of the final hybridization solution to 4.2 μL of Cy3 label, 4.2 μL of Cy5 label, 20 μSSC 3 μ10% SDS 0.6 μL, 95 ° C 2 minutes left in the dark. After standing at room temperature for 3 minutes in the dark, it was spotted at 12 μL / array on a BSA-blocked cDNA microarray. After promptly covering with a 24 mm X 32 mm cover glass (manufactured by Matsunami Glass), it was set in a high pre-cassette (manufactured by Telechem, Hitachi Software, etc.) and allowed to stand at 55 ° C. with light shielding. The cDNA microarray was then removed from the hybrid cassette and immersed in a 2X SSC / 0.1% SDS solution to remove the cover glass. It was immersed in a 2 X SSC / 0.1% SDS solution at room temperature for 20 minutes in a light-shielded manner, and further immersed in a 0.2 X SSC / 0.1% SDS solution for 20 minutes in a light-shielded manner. After immersion in a 0.2 X SSC / 0.1% SDS solution at 42 ° C twice for 20 minutes, the plate was washed with 0.2 X SSC / 0.1% SDS solution and 0.05 X SSC solution. After drying using a centrifuge, detection was performed using an Agilent Microarray Scanner. PMT value indicating detection sensitivity is 100% It was. Axon Instruments' GenePix is used to quantify the fluorescence value of each spot from the scanned image.

Pro version 4.0.1.17 was used.

[0095] BSA blocking of the cDNA microarray is performed at 42 ° C for 45 minutes in a 1% BSA blocking solution (1% BSA, 4 X SSC, 2% SDS) filtered with 0.1 or 0.22 μm of finolet. I went there. After washing twice in water and drying using a centrifuge, it was used for high lysis.

[0096] (4) Selection of effective genes for analysis

The following data cleansing was performed on the expression level data of each gene mounted on the cDNA microarray, and the genes used for analysis were selected.

[0097] a) In the digitized expression data of cDNA microarrays, there is a flag (flag) indicating the effectiveness of the spot (gene) on each array, and the flag indicates “0〃 (valid)”. Selected.

[0098] b) Genes showing expression levels below the average expression level of lambda DNA of the negative control gene plus 2 standard deviations were excluded because they were judged as noise data. The data of the array that can be analyzed after selection in a) and b) is not used for the analysis.

[0099] c) Four individual forces When the correlation coefficient between the two extracted individuals is calculated, and the correlation coefficient power is less than two times out of the six combinations in total, it is regarded as data with poor correlation and the analysis pair Elephant power was also excluded.

[0100] d) Welch t test between the signal value (n = 4) of the sample administration group and the signal value (n = 4) of the vehicle control group (control) in order to select genes with little variation for each experiment. Welc h's t-test) and selected genes with p ≤ 0.05 (null hypothesis: signal value in sample administration group (n = 4) and gene expression level in vehicle control group (n = 4) (signal value) ) Is no difference,).

[0101] e) Threshold values were calculated based on power to select genes with an average expression ratio of n = 4, which were statistically significant. As a result, if the LogRatio (log-transformed ratio of the expression level in the compound-administered group and the expression level in the vehicle control group) was 0.8 or more apart from the reference 0, it was confirmed that the expression ratio was significant. LogRatio = | 0.8 | was set as the threshold value, and the gene whose logRatio fluctuated between 0.8 and 0.8 was selected. [0102] As a result of data cleansing according to the above-e), the number of genes effective for the analysis was 2844. The nucleotide sequences complementary to the partial nucleotide sequences of these genes are shown in SEQ ID NOs: 1 to 2844.

[0103] Further, Tables 5 to 45 show the gene numbers (UniGene) of the genes having base sequences complementary to the base sequences shown in SEQ ID NOs: 1 to 2844 and the clone IDs assigned to the respective base sequences. .

[0104] [Table 5]

[9 挲] [SOTO]

s 拏

8T99lC / 900Zdf / 13d z C06Cl0 / .00i OAV Table 6

7] Table 7

] Table 8

9] [Οΐ 挲] [60Ϊ0]

6 拏

l99l £ / 900Zdf / X3d 83 £ 06CZ0 / Z.00r OAV Table 1 0

1] m [τπο]

I T 拏

T99l £ / 900Zdf / X3d OS εο6εζο / ζ.οοζ OAV βπο]

T99l £ / 900Zdf / 13d e06 € I0 / L00I O Table 1 3

Four] Table 1 4

Five] m [sno]

T99lC / 900Zdf / X3d εοόεζο / ζ, οοζ OAV

T99lC / 900Zdf / X3d εο6εζο .οοε O Table 1 7

8] Table 1 8

9]

6 I

T99lC / 900Zdf / X3d 8S C06CZ0 / .00Z OAV Table 2 0

1]

l99l £ / 900Zdf / 13d 0 £ 06 £ Z0IL00Z O L06 ££ 89SI 8 oi 'ua 9ΐ9εε

Λ 688EC im iti 'ua 6281

8998 998SS 99 £ I 96, 'u U322

8 l9 "UH S98SS 99CI 69 t'u 9952C

098CS I nz \

S9S1 9 SSC \

9S101Z2S ££ 6 丽 'n S S mi

1981 mi οΐδεε 09SI zz

808CC 69 £ 1 \ zn a 'U 89SI 609εε \

n 1921 z z 6121

OOOi 'ua 9921 9989 εε 8121

S9C1 i! SI

909 £ 'u m \ ΙίΗ2 91C1

89 SSSl 9189 -u¾ SlSl

igoif 29 1 in

0i6iru¾ 19CI U8 'u¾ ειει

\ z 09CI 8 (^ m zin

OOiSS 666S01 -UH L ^ Π81

z 01C1

OSAOOI ~ in \ 8 腦 60Π

069εε \ ZZKZ 80S1

889εε s si LOU z z m \ 90SI

O 899εε \ I818E'UH torn 90Π

90 S m \

18SC6 "u¾ 6 S m \ ma C0CI

8 60I 'u¾ \ mz 0 ^ 1 z

19019 'u¾ 6CS1 ιοει e iiru¾ I ε οοει

ζεει 09m 6621

Z 9Cl "u¾ mzz 9C8I 86ZI

0 I 'u¾ 9εει m \

z \\ 8H6'ua H222 96SI

\ 20999 'UH Dragon ¾ ιεει I9Sil "UH m \

αι く一二 GI ： — ニて ^ 泰 a l99TC / 900Zdf / X3d εοόεζο / .οοζ OAV m ιζζιο]

T99l £ / 900rdf / X3d zv εο6ε oo OAV

T99TC / 900Zdf / 13d C06fl0 / .00Z O m [esTo]

ss 拏 l99TC / 900Zdf / X3d εο6εζο / .οοζ; OAV Table 26

27] Table 2 7

8] Table 2 8

9] [OS 挲] [62 TO]

6 S 拏

T99lC / 900Zdf / X3d 817 C06CZ0 / .00Z OAV [οετο]

o ε 拏

l99TC / 900Zdf / 13d 6 eo6czo / .ooz OJW Table 3 1

2] [εε 挲] βειο]

z ε 拏 T99lC / 900Zdf / X3d 1-9 εο6εζο / .οοζ OAV mm [εειο]

£ ε 拏

8l99l £ / 900idf / 13 «I 39 εοόειο / ιοοζ OAV [ss 挲] [^ ετο]

^ ε 拏

T99lC / 900Zdf / X3d 89 εο6εζο / .οοζ OAV [9ε 挲] [ssio]

l99TC / 900Zdf / X3d £ 06 £ WL00Z OAV

9 ε 拏

T99lC / 900Zdf / X3d 99 εο6εζο / .οοζ OAV

L ε 拏 T99lC / 900Zdf / X3d 99 εο6εζο / .οοζ OAV [6ε 挲] [8ε TO]

8 ε 拏

T99lC / 900idf / X3d Z9 C06CZ0 / Z, 00∑; Table 3 9

40] [挲] [0W0]

0 拏

T99lC / 900Zdf / X3d 69 εοόεζο / .οοζ OAV zrni [ΪΟ]

l 拏

8l99lC / 900Zdf / X3d 09 £ 06 0 leg Z ΟΛ / 8ϊ99ϊε9002 /: τ1 £ OAV

[0142] (5) Classification of carcinogens

39 carcinogens (excluding the carcinogens di (2-ethylhexyl) phthalate and phenobarbital shown in Tables 1 to 4) and 20 non-carcinogens (non-carcinogens shown in Tables 1 to 4) The carcinogenicity of the chemical substance was predicted in the following procedure for the 28-day administration group.

[0143] A gene set having a significant difference in expression level between the carcinogen group and the non-carcinogen group was selected based on Welch's t-value. The expression pattern of the selected gene set was classified using the cluster analysis function of Gene Maths (Applied Maths, Sint-Martens- Latem, Belgium) to determine the shape force cluster of the saddle diagram. Change the Welch t-value selection conditions, confirm the cluster structure of the substance with several patterns of gene sets V, group 1 of two carcinogens that stably form the same cluster even if the gene set is changed, 2 was selected.

[0144] Table 43 shows the gene set selection conditions (Welch t-value) and the clone ID of the selected gene. In addition, Figures 1 to 6 show saddle diagrams that were clustered using the gene set selected for each Welch t value. Figures 1, 3, and 5 are part of the saddle plots obtained by cluster analysis using gene sets selected under conditions of | t values | ≥2.0, 2.3, and 2.5, respectively. Figures 2, 4, and 6 are the remainder of the saddle view shown in Figures 1, 3, and 5, respectively. In Figures 1-6, 參 represents a carcinogen and ◯ represents a non-carcinogen.

[0145] [Table 43]

Table 4 3

Next, using the remaining carcinogens excluding carcinogen groups 1 and 2 and all non-carcinogens, the substance groups were selected in the same way using the same expression patterns that exist in the remaining carcinogens. That is, the genes that varied characteristically between the remaining carcinogen-administered group and all the non-carcinogen-administered groups were selected based on the Welch t value, and then cluster analysis was performed using the gene set. We changed the Welch t-value selection conditions, confirmed the cluster structure of the substances using several gene sets, and selected carcinogen group 3 that always forms clusters stably. Table 44 shows the gene set selection conditions (Welch t-value) and the clone ID of the selected gene. In addition, Figs. 7-9 show the saddle diagrams that were cluster-analyzed using the gene set selected for each Welch t value. Figures 7, 8 and 9 show the t values | ≥Diagrams obtained by cluster analysis using gene sets selected under conditions of 2.3, 2.5, and 2.8. In Figures 7-9, 參 represents a carcinogen and ◯ represents a non-carcinogen.

[Table 44]

Table 4 4

The chemical substances classified into carcinogen groups 1 to 3 are as follows.

(Group 1) Carbon tetrachloride, jetyl-tolosamine, 2-tropropane, N-nitrosomorpholine, ethynylestradiol, safrole, N-nitrosopiperidine, acetamide, jetylstil bestronore

(Group 2)

Clofibrate, _α -hexachlorocyclohexane, trichlorethylene, ニトロ -nitrosodimethylmethylamine, 2-amino-3,8-dimethylirimidazo [4,5-f] quinoxaline, 2-amino-1-methyl-6-phen- Rimidazo [4,5-b] _pyridine, Benz (a) Anthracene, 7, 12-Dimethylbenzanthracene, 3-Methylcholanthrene, 4-Nitroquinoline-1-oxide, N-Ethyl-N-nitrosourea Tannic acid, urethane, sodium phosphate salt, D, L-ethionine

(Group 3)

Trichloro oral acetic acid, pentachloroethane, black mouth form, benzo ( _a ) pyrene, N-methyl-Ν '-to --- nitrosoguanidine, tetrachloroethylene

(6) Prediction formula creation procedure and verification method

Before creating the prediction formula, the target carcinogen group and non-carcinogen group were determined. The carcinogen group was selected from carcinogen groups 1-3. The non-carcinogenic substance group used all non-carcinogenic substances that can be analyzed.

[0149] Welch t-values are gene sets that show characteristic variation between the selected carcinogen group and non-carcinogen group, and the absolute value of the average LogRatio of each carcinogen group is the non-carcinogen group. It was selected under the condition that it was larger than the absolute value of the average value of LogRatio of the group.

[0150] Using Support Vector Machine (SVM) and free software SVMlight (based on URL http://svmlight.joachims.org/) created based on supervised classification method, follow the steps below A carcinogenicity prediction formula was created in

[0151] a) If the number of substances belonging to the carcinogen group and the non-carcinogen group is not equal, perform random sampling for the number of the two substances with the smaller number of substances in both groups, and calculate the number of substances in both groups. It was equal.

[0152] b) Using both substance groups with the same number of substances and the selected gene set, we learned with SVMlight and created a prediction formula. [0153] c) Carcinogenicity prediction of both substance groups (training set) used to create the prediction formula to verify the effectiveness of the prediction formula, and leave one of the substances that is one of the verification methods Cross validation by Leave one out was performed.

[0154] d) Prediction of the remaining substance data (test set) not used in creating the prediction formula.

[0155] e) The selection conditions were adjusted until the gene set showing the highest prediction rate was obtained, and the procedures from a) to were repeated.

[0156] (7) Predictive formula creation for each expression pattern

Carcinogenicity prediction formula for carcinogen group 1 (0 was created according to the procedures in (6) a). Table 45 shows the results of prediction of the training set and test set using the prediction formula (0) for the constructed group 1.

[0157] [Table 45] Table 4 5

[0158] As a result, the training set was 100% predictable. Prediction formula (0 was confirmed to be able to predict group 1 of unknown carcinogens with a probability of 89.7% by L00 cross-validation. Also, carcinogens in the test set could hardly be predicted, and non-carcinogens Was 81.8%, and this result shows that the prediction formula (0 is the fluctuation pattern of group 1) It is a special prediction formula, and carcinogen groups 2 and 3 are thought to have been unable to predict due to different fluctuation patterns.

[0159] Similarly, for Carcinogen Group 2 and Group 3, the prediction formula GO and (iii) were created.

. Tables 46 and 47 show the results of predicting each training set and test set using each prediction formula.

[0160] [Table 46] Table 4 6

[0161] [Table 47]

Table 4 7

[0162] As with the results of Group 1, the prediction results for Groups 2 and 3 were relatively high for the non-carcinogens in the training set and test set. As a result of L00 cross-reduction, it was confirmed that Group 2 and Group 3 could be predicted with a probability of 82.9% for the prediction equation GO and 84.6% for the prediction equation (iii). As in the case of Group 1, the prediction formulas (ii) and Gii) are predictive formulas specific to the substance group, so the carcinogenicity of the test set was unpredictable.

[0163] (8) Construction of prediction flow

By combining the prediction formulas for each carcinogen group created in (7), a prediction flow for predicting carcinogens with different gene expression patterns was constructed (Figure 10). As a result of applying the data of 59 substances used in creating the prediction formula to the prediction flow, the prediction formula (0 was able to predict carcinogenicity ¾ substance, but two non-carcinogenic substances were incorrectly predicted as carcinogens. Prediction formula (excluding 11 substances that were predicted to be carcinogens at 0!) For the remaining substances! As a result of predicting by the prediction formula GO, 27 carcinogens could be predicted, but 5 non-carcinogens. As a result of predicting the remaining substances using the prediction formula (iii) except for 22 substances that were predicted to be carcinogens using the prediction formula GO, one carcinogen could be predicted. Two non-carcinogens were incorrectly predicted as carcinogens. There were two carcinogens that could not be predicted in the end.

[0164] As a result of combining the prediction formulas created for each carcinogen group with patterns characteristic to carcinogens and performing carcinogenicity prediction using the prediction flow, 37 carcinogens could be predicted.

[0165] Example 2 (carcinogenicity prediction by decision tree)

The expression data obtained in (1) to (4) of Example 1 were analyzed using a decision tree

[0166] In performing analysis using a decision tree, the following numerical conversion was performed on the data after data cleansing.

[0167] A value (Log Ratio) obtained by converting the ratio (Ratio) between the expression level of the compound administration group and the expression level of the vehicle control group into Log was calculated. When LogRatio ≥ 0.8 (0.8, which can be said to have a significant change in LogRatio), it was considered that the expression level had increased, and all were converted to 1. Similarly, when LogRatio ≤_0.8, the amount of expression was considered to have decreased, and all were converted to -1. In all other cases, it was assumed that there was no change and converted to 0.

[0168] The decision tree was analyzed using Classification and Regression Trees Ver.5 (Salford Systems). Figure 11 shows the results of analysis under the following conditions.

• Substances to be analyzed 40 carcinogens, 21 noncarcinogens

• 897 genes remaining after gene set data cleansing

'Branch algorithm Gini branch rule is adopted

'10 -fold Cross Validation

In FIG. 11, “Node” indicates a branch point, and “Terminal node” indicates a terminal branch point. “ID” is the clone ID number, “W” is the number of substances classified at the branch point, and “TS” is the substance number classified at the terminal branch point. Substance numbers with underline indicate carcinogens, underlined! Indicate non-carcinogens.

[0169] As shown in Fig. 11, when the gene expression level of ID: 35953 selected at branch point 1 is increased (branch point 7), only carcinogens are classified at terminal branch point 7, and terminal branching is performed. At point 8, substances from both categories were mixed. In addition, gene expression level decreased or did not fluctuate In this case (Tree on the right side of branch point 2), only carcinogens were classified at terminal branch points 1,3,4,6. Of the 9 substances classified as terminal branch point 2, 7 were carcinogens, so terminal branch point 2 was classified as a carcinogen, and 2 non-carcinogens were mistakenly classified as carcinogens. It is thought. At terminal branch point 5, only non-carcinogenic substances were classified. Table 48 shows the genes selected at the seven branch points.

[0170] [Table 48] Table 4 8

[0171] As a result of cross-validation, the saddle diagram constructed with seven branch points could predict the carcinogenicity of 52 substances, excluding 7 substances that could not be distinguished.

[0172] In Examples 1 and 2, in order to calculate the predictive rate of carcinogenicity of chemical substances, saddle diagrams were created for carcinogenic and non-carcinogenic substances. Even when a saddle diagram is created using only carcinogens, carcinogens are classified in the same way as in Example 1 or 2, and the carcinogenicity of the test substance can be predicted.

Claims

The scope of the claims

[1] Multiple carcinogens are administered to each carcinogen-administered group, and mRNA is collected from each carcinogen-administered group after a predetermined period, and the mRNA expression level is measured to significantly increase the mRNA expression level. Alternatively, compare the expression pattern of the expression variation gene selected in the first step and the expression variation gene selected in the first step between the carcinogen administration groups and the expression pattern based on the similarity of the expression pattern. A second step of classifying into a plurality of groups and preparing expression patterns of expression variable genes classified into a plurality of groups;

Method for predicting carcinogenicity of test substance having

[2] Multiple carcinogens were administered to each carcinogen-administered group, and mRNA was collected from each carcinogen-administered group after a predetermined period of time, and the mRNA expression level was measured to significantly increase the mRNA expression level. Alternatively, compare the expression pattern of the expression variation gene selected in the first step and the expression variation gene selected in the first step between the carcinogen administration groups and the expression pattern based on the similarity of the expression pattern. A second step of classifying into a plurality of groups and preparing expression patterns of expression variable genes classified into a plurality of groups;

Change in expression of carcinogen-administered group in which NA was collected and expression pattern of expression-variable gene was obtained, and expression pattern of expression-variable gene of test substance administration group was prepared in the second step in advance A third step of calculating the degree of coincidence in comparison with the gene expression pattern;

A method for predicting the carcinogenicity of a test substance having the following three steps:

Carcinogen administration group strength Fluorescently labeled cDNA or c prepared using collected mRNA Fluorescence pattern obtained by hybridization of RNA to DNA microarray and non-carcinogenic substance administration group administered non-carcinogenic substance Fluorescently labeled cDNA or cRNA prepared using collected mRNA hybridized to DNA microarray Comparing the fluorescence pattern obtained by sizing and selecting the expression variation gene by detecting the significant difference from the fluorescence intensity of each spot of the fluorescence pattern,

[3] DNA microarray ability mRNA is extracted from test animals to which carcinogens have been administered, and the expression level is increased or decreased. All or part of the cDNA sequence synthesized by reverse transcription reaction is amplified or chemically synthesized by mRNA. The method for predicting carcinogenicity of a test substance according to claim 2, which is a DNA microarray on which the synthesized DNA is mounted.

[4] The subject of any one of claims 1 to 3, wherein the expression variable gene is selected from five or more genes having a base sequence complementary to the base sequence described in SEQ ID NOs: 1 to 2844. A method for predicting the carcinogenicity of test substances.

[5] The carcinogens administered to the carcinogen administered group are clofibrate, di (2-ethylhexyl) phthalate, carbon tetrachloride, 2,4-diaminotoluene, quinoline, phenobarbital, jetylnitrosamine, 2 -Nitropropane, N-nitrosomorpholine, aldrin, di-adipate (2-ethinorehexinole), ethinoreest radionore, hexacyclobenzene, _a- hexaclo Rocyclohexane, Trichlorethylene, Butylated hydroxyasol, Limonene, Safrol, 1,4-Dichlorobenzene, 1,4-Dioxane, Furan, Methyl carbamate, Thioacetamide, N-Nitrosodimethylamine, N-Nitroso Piperidine, 2-amino-3,8-dimethylimidazo [4,5-f] quinoxaline, 2-amino-1-methyl-6-fe-louimidazo [4,5-b] -pyridine, benz (a) anthracene, 7, 12-dimethylbenzanthracene, 3-methylcholanthrene, 4-nitroquinoline-1-oxide, N-ethyl-N-nitrosourea, trichloroacetic acid, tannic acid, urethane, pentachloroethane, chloro Mouth form, benzo ( _a ) pyrene, N-methyl --'- nitro-N-nitrosoguanidine, tetrachloroethylene, acetamide, jetylstilbestrol, sodium phenyloline The method for predicting the carcinogenicity of a test substance according to any one of claims 1 to 4, comprising at least four selected from a salt, D, L-ethionine, 4-dimethylaminoazobenzene, and chlorendic acid.

[6] The method for predicting carcinogenicity of a test substance according to any one of claims 1 to 4, wherein the carcinogen administered to the carcinogen-administered group contains at least six carcinogens according to claim 5.

[7] The method for predicting carcinogenicity of a test substance according to any one of claims 1 to 4, wherein the carcinogen administered to the carcinogen administered group contains at least nine carcinogens according to claim 5.

[8] Non-carcinogens administered to the non-carcinogen-administered group are 2,6-diaminotoluene, 8-hydroxyquinoline, D-manntol, L-ascorbic acid, 2-chloroethanol, 2- (chloromethyl) Pyridine hydrochloride, dl-menthol, 4--tro- 0-phenol-diamine, benzoin, odoform, lithocholic acid, lindane, 2-chloro-p-phenol-diamine amine sulfate, p-phenol-diamine amine hydrochloride , 2,5-toluenediamine sulfate, aspirin, 4- (chloroacetyl) acetanilide, phthalamide, force prolatatum, 1-black mouth—2-propanol, 3-chloro mouth—p-toluidine, glutaraldehyde, 4-nitro The method for predicting carcinogenicity of a test substance according to any one of claims 2 to 7, comprising at least one selected from anthrallic acid and 1-tronaphthalene.

[9] The method for predicting carcinogenicity of a test substance according to any one of claims 2 to 7, wherein the non-carcinogenic substance administered to the non-carcinogen-administered group contains at least five non-carcinogenic substances according to claim 8.

[10] The non-carcinogenic substance according to claim 8, wherein the non-carcinogenic substance administered to the non-carcinogenic substance-administered group is The method for predicting the carcinogenicity of a test substance according to any one of claims 2 to 7, comprising at least 10 substances.

[11] The method for predicting carcinogenicity of a test substance according to any one of [1] to [10], wherein the classification of the expression pattern of the expression variable gene in the carcinogen-administered group is performed by cluster analysis or decision tree.

[12] The method for predicting the carcinogenicity of a test substance according to any one of claims 1 to 11, wherein the expression pattern of the expression variable gene is classified into 3 or more groups.

[13] The carcinogenicity of the test substance according to any one of claims 1 to 12, wherein mRNA is collected for each carcinogen-administered group force after 1 to 90 days after administration of the carcinogen to each carcinogen-administered group. Sex prediction method.

[14] The carcinogenicity of the test substance according to any one of claims 1 to 12, wherein after the administration of the carcinogen to each carcinogen-administered group, mRNA is also collected from each carcinogen-administered group force after 14 to 28 days. Prediction method.

[15] The carcinogenicity of the test substance according to any one of claims 1 to 14, wherein mRNA is collected from the test substance administration group 1 to 90 days after the test substance is administered to the test substance administration group Prediction method.

[16] Carcinogenesis of a test substance according to any one of claims 1 to 14, wherein mRNA is collected from the test substance administration group after 14 to 28 days after the test substance is administered to the test substance administration group Sex prediction method.