[go: up one dir, main page]

WO2016193977A2 - Méthodes de prévision de l'hépatotoxicité - Google Patents

Méthodes de prévision de l'hépatotoxicité Download PDF

Info

Publication number
WO2016193977A2
WO2016193977A2 PCT/IL2016/050566 IL2016050566W WO2016193977A2 WO 2016193977 A2 WO2016193977 A2 WO 2016193977A2 IL 2016050566 W IL2016050566 W IL 2016050566W WO 2016193977 A2 WO2016193977 A2 WO 2016193977A2
Authority
WO
WIPO (PCT)
Prior art keywords
genes
gene
panel
compound
expression
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/IL2016/050566
Other languages
English (en)
Other versions
WO2016193977A3 (fr
Inventor
Ganit YARDEN
Assaf Wool
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Compugen Ltd
Neviah Genomics Ltd
Original Assignee
Compugen Ltd
Neviah Genomics Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Compugen Ltd, Neviah Genomics Ltd filed Critical Compugen Ltd
Publication of WO2016193977A2 publication Critical patent/WO2016193977A2/fr
Publication of WO2016193977A3 publication Critical patent/WO2016193977A3/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/5044—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics involving specific cell types
    • G01N33/5067—Liver cells
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/5014—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics for testing toxicity
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/502—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics for testing non-proliferative effects
    • G01N33/5023—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics for testing non-proliferative effects on expression patterns
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/142—Toxicological screening, e.g. expression profiles which identify toxicity
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/158—Expression markers

Definitions

  • DILI Drug-induced liver injury
  • ALT alanine aminotransferase
  • AST aspartate aminotransferase
  • AP alkaline phosphatase
  • GTT ⁇ -glutamyl transpeptidase
  • elevated plasma levels of these enzymes can also be caused by extra-hepatic injuries.
  • Other common markers include total bilirubin (hepatobiliary injury) and serum bile acids (general liver damage) which are also influenced by hemolysis and diet, respectively.
  • HCS High content screening assays that simultaneously measure multiple end points and associated cellular reactions offer an opportunity to detect changes of DILI related molecular signals in live cells with higher sensitivity.
  • the HCS assays can only offer limited sensitivities to identify liver drug induced injury due to the complexity of toxicity mechanisms.
  • end point measurements are based mostly on visible intracellular changes that provide useful information only when the toxic process is in relatively advanced stage, demonstrates a need in the art for an assay to determine toxicity of an unknown compound earlier in the screening process.
  • Toxicogenomics-based models enable unprecedented opportunities to comprehensively assess alterations in gene-expression profiles induced by drugs.
  • Several reports have illustrated the value provided by toxicogenomics to confirm various mechanisms of liver toxicity in rats, such as changes in fatty acid ⁇ -oxidation and lipid metabolism, glutathione depletion, disruption of mitochondrial homeostasis and ATP production, peroxisome proliferation, induction of cytochromes P450 and Phase II metabolism enzymes, hypoxia, oxidative stress, or proteosomal changes (Boess et al. (2007) Toxicol. In Vitro, 21 :1276-86; Dai et al. (2006) Genome Inform., 17:77-88; Jolly et al. (2004) Toxicol.
  • These studies use global gene expression analyses to detect changes that influence, predict, or help define drug toxicity.
  • By evaluating and characterizing differential gene expression after exposure to drugs it is possible to use complex expression patterns to predict toxicologic outcomes and to identify mechanisms involved with or related to the toxic event. This leads to a better understanding of the mechanisms of toxicity by the identification of toxicity stages.
  • a growing number of studies demonstrate that gene expression data are useful for class prediction studies, in which expression signature from known toxins are used to predict the toxicological class of an unknown toxicant (e.g. new chemical entity or drug candidate).
  • gene expression data can provide an early indication of toxicity because toxin mediated changes in gene expression are often detectable before clinical chemistry, histopathology, or clinical observations suggest a toxic effect.
  • hepatic toxicity is traditionally identified through the integrated evaluation of histopathological findings and clinical pathology parameters, such as serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST).
  • ALT serum alanine aminotransferase
  • AST aspartate aminotransferase
  • this approach is relatively robust to detect hepatotoxicity.
  • changes in these parameters may occur only after prolonged periods of dosing, or sometimes respond only to short-term, acute exposure such as acetaminophen overdose, and may be subtle and challenging to interpret in early, short-term studies. Therefore, additional methods are needed for the early detection of hepatotoxicity.
  • DILI prediction is difficult primarily for chronic toxicity (compared to acute toxicity) which is associated with prolonged exposure in low concentrations of drug.
  • the limited power of DILI prediction is mostly attributed to the complex nature of DILI, a poor understanding of mechanisms, a scarcity of human hepatotoxicity data and insufficient bioinformatics capabilities.
  • Major investments have therefore been made to reduce risk for DILI, including the search for, and validation of predictive genomic biomarkers.
  • Drug-induced toxicity gene biomarkers are aimed at assisting in the early detection of toxic drug effects, allowing early drug prioritization and a cost-effective drug development process, whereby drug candidates having high risk for toxic effects will not progress into more expensive drug development stages.
  • genomic biomarkers are frequently more sensitive than the traditional, functional and morphological markers, they represent a useful supplement to more reliably detect hepatotoxicity in short-term studies, where the full phenotypic manifestation of toxicity may not have fully developed.
  • the improved sensitivity of genomic biomarkers is particularly valuable when used in the context of short-term exploratory studies for prioritization and selection of potential drug candidates. This is especially evident for toxic changes for which current toxic biomarkers are not very sensitive or are lacking.
  • this invention provides a method of predicting hepatotoxicity of a compound, said method comprising the steps of:
  • said panel of genes further comprises gene F3, Lgals2, Ltbp4, SerpinO, C4bpb, Cbs, Dapkl, Fblnl , Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the hepatocytes used in a method of this invention comprise primary hepatocytes.
  • methods of this invention use of a low density array.
  • comparing gene expression comprises correlating the gene expression with an amount of signal or change in signal.
  • predicting comprises detecting if the expression of genes is up-regulated or down-regulated in the culture system in comparison with a negative or positive control culture system, or if the gene expression is substantially identical in the culture system and a positive control culture system.
  • this invention provides a use of a panel of genes comprising Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn as marker genes for screening a compound for hepatotoxicity activity.
  • the panel of genes further comprises gene F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • this invention provides use of nucleic acid probes specifically hybridizing under stringent conditions with a panel of genes comprising Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn, or gene products encoded by said panel of genes, or respective parts thereof, for detecting and quantitating the expression of said panel of genes, wherein said panel of genes is representative for a hepatotoxic activity in a cellular system.
  • the panel of genes further comprises gene F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • this invention provides an array for screening a compound for hepatotoxic activity, comprising nucleic acid probes that are capable of specifically hybridizing under stringent conditions with a panel of genes, said panel of genes comprising Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn, or gene products encoded by said panel of genes, or respective parts thereof.
  • the array is a low density array.
  • the panel of genes further comprises gene F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • this invention provides a method of analysis of toxicity prediction data, comprising (a) training a classifier based on the expression of the signature genes in experiments with compounds of known toxicity; (b) calculating a single toxicity score for a gene or group of genes; and (c) defining a toxicity threshold value based on a range of toxicity scores.
  • Figures 1A and IB presents distribution of expression values (loglO) for two representatives Illumina bead-chip arrays.
  • Figure 1A shows a comparison between two samples which were treated by the same compound (a biological repeat; sample 1 solid line, sample 2 dashed line).
  • Figure IB presents scatter plots comparing the expression values (loglO) of two Illumina bead-chip arrays.
  • the solid grey line represents the same expression for both arrays, while the dashed grey lines represent a factor of 2 difference in expression above or below the equality line.
  • the scatter plot compares two samples treated by different compounds.
  • Figures 2 A and 2B present normalized expression values of Illumina bead chip arrays for representative genes.
  • the y axis represents the compounds, The circles are arranged in lines/rows, each row contains the expression values of one gene for several samples treated by the same compound. This compound's abbreviation appears on the left. The stippled line separates toxic and non-toxic compounds.
  • ROSI Rosiglitazone
  • EMD EMD l-(2-trifluoromethoxyphenyl)-2-nitroethanone
  • Fenofibrate Fenofibrate
  • FENO Fenofibrate
  • TROG Troglitazone
  • CHLO Chlorpromazine
  • CYCL Cyclosporine A
  • AIO Amiodarone
  • VAL Valproic acid
  • Metformin MetF
  • Methanol MET
  • Ibuprofen IBU
  • Dimethyl sulfoxide DMSO
  • Figures 3A-3D Figures 3 A-3D.
  • Figure 3 A left panels: Light microscopy images of primary rat hepatocyte sandwich cultures growing in a serum free media for the indicated time points after seeding (d-day). Defined cellular membranes, constant nuclear size and clear cytoplasm were observed until day 14 after seeding.
  • Figure 3A right panels: Accumulation of carboxy-DCFDA (5-(and-6)-Carboxy-2',7'-Dichlorofiuorescein Diacetate) in bile canalicular-like structures of serum- free primary rat hepatocyte sandwich cultures at the indicated time points after seeding, (d- day).
  • carboxy-DCFDA 5-(and-6)-Carboxy-2',7'-Dichlorofiuorescein Diacetate
  • Figure 3B Fold change in expression levels of P450 enzyme genes CYP1A, CYP2B, CYP2C, and CYP3A, as measured by qRT-PCR using TaqMan primers. Ct values were normalized using the housekeeping gene Rpl37A.
  • Figure 3C Induction of EROD (7-ethoxy-resorufin-O- deethylase) activity (CYP1A), BROD (7-Benzyloxyresorufin-O-deaikylase oxidation) activity (CYP2B), P450-Glo CYP2C, and P450-Glo CYP3A activity. Induction performed as described in Figure 3B. Change in enzyme activities are displayed relative to the control cells (no induction). Standard error of the mean values was below 33% for each parameter and error bars were omitted to maintain clarity of the results presentation. Induction of P450 enzyme activity and expression during the cell growth period suggest the hepatocytes are metabolicaily active.
  • Figure 3D Secretion of Albumin by primary rat hepatocyte sandwich cultures growing in a serum free media after the indicated time points after seeding. The amount of albumin secreted by the cells per hour was measured in the cells conditioned media using the Albumin Rat Eliza Kit (Abeam). Results are shown as mean values ⁇ standard error, from 3 independent experiments. The amount of albumin secreted from the cells was stable over the period of growth (16 days), suggesting for a functional hepatocytes.
  • Figures 4A-C Scatter plots comparing normalized TaqMan Low Density Arrays (TLDA) expression (Ct) and normalized bead chip expression (loglO; Merck-Serono data of Example 1) for three genes.
  • Figure 4A shows a scatter plot for Habp2, illustrating Habp2 gene expression correlation with toxicity.
  • Figure 4B shows a scatter plat for Dapkl, illustrating Dapkl gene expression correlation with toxicity.
  • Figure 4C shows a scatter plot for Gale, illustrating Gale gene expression correlation with toxicity. Each point is one sample, empty circles for nontoxic compounds, filled circles for toxic compounds and grey filled circles for control samples. Results show a good correlation in Figure 4A and Figure 4B, and a weak correlation in Figure 4C.
  • Figures 5A-5C Normalized expression levels for one gene (Acot3) measured by qRT- PCR in TaqMan low density array (TLDA).
  • Figure 5A shows samples obtained from an outside source.
  • Figures 5B and 5C show samples produced based on the Description and Examples herein below, after 7-day and 14-day treatment times, respectively.
  • Compounds tested were: Rosiglitazone (ROSI), Chloropromazine (CHLO), Cyclosponine A (CYCL), Amiodarone (AMIO), Metformin (METF), and Dimethyl sulfoxide (DMSO). Each point is a single sample. Empty circles are for non-toxic labeled compounds and control samples, filled circles are for toxic labeled compounds.
  • ROSI Rosiglitazone
  • CHLO Chloropromazine
  • CYCL Cyclosponine A
  • AMIO Amiodarone
  • MEF Metformin
  • DMSO Dimethyl sulfoxide
  • Figures 6A and 6B Normalized expression levels of the C6 gene ( Figure 6A), and Rpl37 ( Figure 6B), from the training experiment. Each point represents a single sample.
  • Compounds used for training were: Ximelagatran (XIME), Tolcapone (TOLC), Rosiglitazone (ROSI), l-(2-trifluoromethoxyphenyl)-2-nitroethanone (EMD), Fenofibrate (FENO), Troglitazone (TROG), Chloropromazine (CHLO), Cyclosporine A (CYCL), Aceteminophen (ACET), Amiodarone (AMIO), Valproic acid (VAL), Pioglitazone (PIOG), Entacapone (ENTA), Buspirone (BUSP), Isomer of Acetaminophen (AMAP), Metformin (METF), Methanol (MET) , Ibuprofen (IBU), and Dimethyl sulfoxide (DMSO). Empty circles are for non-
  • FIG. 7 Gene Selection Flowchart describing the process of the classifier's genes selection.
  • the gene numbers presented are merely exemplary. Absolute gene numbers may differ dependent on the starting pool.
  • Gene selection steps include: The process started with gene expression measurements for all transcribed genes, from experiments where rat primary hepatocytes were exposed to toxic and non-toxic compounds at high and low dosage levels as well as to DMSO 0.5%, as a negative control. .
  • the first step is removing all genes with an expression level below a threshold expression level defined as loglO value of 2.2 (7000).
  • genes useful for deciding toxicity are selected, wherein two definitions of toxicity were used, wherein definition one does not consider the expression results of low dose toxic compounds (7001b), and definition two considers the low dose toxic compounds to be non-toxic (7001a).
  • a classifier is trained to predict toxicity using the expression of the genes still in the list, measuring the importance of each gene to the training. This process is repeated many times and genes are ranked according to their average importance score. The 300 highest ranking genes based on the classifier results (7003a and 7003b) are selected for further evaluation.
  • the gene list of step 7003a and 7003b are merged (7005).
  • the next step removes genes with a median expression differences between toxic and non-toxic compounds of less than 0.1, or of compounds with high toxic and non-toxic distribution overlaps (7007).
  • manual gene selection was used to evaluate the remaining genes present in the pool. Evaluation criteria included: comparing expression with other data bases showing expression under toxic conditions, literature evaluation, relevance of gene pathways, and expression in other tissues than liver (7009). Based on the gene selection parameters and starting pool of 22,523 genes from Example 1 , the flowchart shows a resultant final list of 60 genes were selected.
  • FIG. 8 The Full Toxicity Prediction Process Flowchart describes a full process of toxicity prediction by the classifier.
  • step 8001 primary hepatocytes are isolated and viability tested using Trypan blue staining.
  • QAM Quality Assurance step 1
  • hepatocyte isolations with viability less than 84% are discarded, and isolations with at least 84% viability proceed to the next step of hepatocyte exposure to a "test" compound (8005).
  • Hepatocytes cultured, in one embodiment, in a collagen sandwich culture are analyzed for functional and morphological properties.
  • RNA and cDNA are prepared from the cultured hepatocytes.
  • QAM Quality Assurance step 3
  • the total RNA isolated is analyzed using the RNA Screen Tape 2200 TapeStation instrument and software (Agilent Technologies, CA, USA).
  • RNA samples with a RIN above 8 are used for quantitative Polymerase Chain Reaction on TaqMan-Low Density Arrays (qPCR TLDAs) (8013).
  • the gene expression results are then analyzed using the classifier software (8015), providing an answer if the "test" compound is toxic or not (8017).
  • Figure 9 Correlation between expression levels of Vtn gene and Cyp2c6 gene in samples exposed to different compounds. Each symbol represents a different tested sample. "Circles”- DMSO samples; “Squares”- samples treated with non-toxic compounds; “Xs” - samples treated with toxic compounds. [0024] Figure 10. An example of a way of clustering compounds into groups according to the gene expression behavior, as it may be performed for any new compound that is tested. The data presented shows clustering of the compounds tested in Example 7. DETAILED DESCRIPTION OF THE PRESENT INVENTION
  • This invention provides in one embodiment, a method to predict hepatotoxicity of a compound by providing an in vitro hepatocyte culture system capable of expressing a panel of genes, wherein the culture may be incubated with the compound to be screened and the expression level of the panel of genes measured, and wherein comparing the expression level of the panel of genes in the culture system exposed to the compound being screened with the gene expression of the same panel of genes in a control culture system, predicts the hepatotoxicity of the compound.
  • the terms "hepatotoxicity” or “hepatotoxic” may be used interchangeably with "toxicity” or "toxic” as it refers to the properties of a compound(s) being screened, having all the same qualities and meaning.
  • this invention provides a process of computationally constructing a panel of genes for classifying unknown compounds as having hepatotoxic activity. In another embodiment, this invention provides a process of computationally constructing a panel of genes for classifying unknown compounds as not having hepatotoxic activity. In another embodiment, a subset of a panel of genes may be used for classifying unknown compounds as hepatotoxic or as not hepatotoxic. [0028] In one embodiment, this invention provides a use of a panel of genes comprising marker genes for screening unknown compounds for hepatotoxic activity. In another embodiment, a percentage of the panel of marker genes may be used to predict the hepatotoxicity of an unknown compound.
  • this invention provides use of nucleic acid probes that specifically hybridize under stringent conditions with a panel of genes, or gene products encoding the panel of genes, or respective parts thereof, for detecting and quantitating the expression of the panel of genes, wherein the panel of genes represents hepatotoxic activity in a hepatocyte cellular system.
  • nucleic acid probes used may also include those that specifically hybridize under stringent conditions with control genes.
  • this invention provides an array for screening a compound for hepatotoxic activity, wherein the array comprises nucleic acids probes that specifically hybridize under stringent conditions (60°C, lmin, 45 cycles) with a panel of genes, or gene products encoding the panel of genes, or respective parts thereof, for detecting and quantitating the expression of the panel of genes, wherein the panel of genes represents hepatotoxic activity in a hepatocyte cellular system.
  • the array is a low density array.
  • stringent conditions refers to conditions that are compatible to produce binding pairs of nucleic acids, e.g., probes and targets, of sufficient complementarity to provide for the desired level of specificity in the assay while being incompatible to the formation of binding pairs between binding members of insufficient complementarity to provide for the desired specificity.
  • Stringent assay conditions are the summation or combination (totality) of both hybridization and wash conditions.
  • stringent hybridization in the context of nucleic acid hybridization are sequence dependent, and are different under different experimental parameters.
  • stringent conditions comprise incubating at 50 to 80°C for 10 to 20 hours.
  • stringent conditions comprise incubating at about 65°C for 10 to 20 hours.
  • Advantages for using an array in the screening process include running concurrent tests of portions of a hepatocyte culture, wherein data may be provided for response of thousands of genes to different known toxic or known non-toxic compounds.
  • a method to predict hepatotoxicity of a compound comprises the steps of: (a) providing a hepatocyte culture system capable of expressing a panel of genes comprising genes Cyp2c6vl , F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 or Vtn; (b) incubating the hepatocyte culture with said compound; (c) measuring the gene expression of said panel of genes following the incubation step (b); (d) comparing gene expression of said panel of genes, with gene expression of a control hepatocyte culture system; and (e) determining hepatotoxicity based on said comparison of gene expression.
  • a method to predict hepatotoxicity of a compound comprises the steps of: (a) providing a hepatocyte culture system capable of expressing a panel of genes comprising genes Cyp2c6vl , Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn; (b) incubating the hepatocyte culture with said compound; (c) measuring the gene expression of said panel of genes following the incubation step (b); (d) comparing gene expression of said panel of genes, with gene expression of a control hepatocyte culture system; and (e) determining hepatotoxicity based on said comparison of gene expression.
  • the panel of genes further comprises gene F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • Hepatotoxicity also called liver toxicity or drug-induced liver injury (DILI) refers to a chemically-induced or driven liver damage.
  • the liver plays a central role in the metabolism of various endogenous and exogenous chemicals in the body. Due to its function as a central metabolizing organ, the liver is, in particular, susceptible to the toxicity from toxic agents or their metabolites.
  • Chemical compounds that may prove toxic to the liver include drugs administered as part of a medical treatment, nutritional compounds contained by food as well as chemical compounds taken up from the environment. Many compounds may be chemically modified by the liver in a two-step process, wherein the resultant product or intermediate metabolite may prove to be hepatotoxic.
  • the first step usually, comprises functionalization of the compound.
  • the functionalized compounds will be conjugated in order to allow for, e.g., excretion via the bile.
  • a gene is referred to herein in one embodiment, as a region on the genome that is capable of being transcribed to RNA that either has a regulatory function, a catalytic function and/or encodes a protein.
  • a gene typically has introns and exons, which may organize to produce different RNA splice variants that encode alternative versions of a mature protein.
  • the term "gene” includes fragments of genes that may or may not represent a functional domain.
  • a "panel of genes” as used herein refers in one embodiment, to a group of identified genes whose levels of expression vary in response to known hepatotoxic and non-hepatotoxic compounds, as measured in an in vitro hepatocyte culture under different conditions.
  • the term "panel of genes” may be used interchangeably with the term “plurality of genes” having all the same meanings and qualities.
  • the different conditions may be caused by exposure to certain agent(s) - whether exogenous, endogenous, synthetic or a natural product- at differing concentrations and for different amounts of time. The expression of a plurality of genes demonstrates certain patterns.
  • each gene in the plurality is expressed differently under different conditions or with or without incubation with a range of products (exogenous, endogenous, synthetic or a natural product, or metabolites or derivatives thereof).
  • the extent or level of differential expression of each gene may vary in the plurality and may be determined qualitatively and/or quantitatively according to this invention.
  • a gene expression profile refers to a plurality of genes that are differentially expressed at different levels, which constitutes a "pattern” or a "profile.”
  • the term "expression profile”, “profile”, “expression pattern”, “pattern”, “gene expression profile”, “gene expression” and “gene expression pattern” are used interchangeably having all the same meanings and qualities.
  • a panel of genes of the methods, uses and arrays described herein further comprises at least an additional gene.
  • the panel of genes comprises genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the panel of genes further comprises gene F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the panel of genes comprises genes Cyp2c6vl , F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl , Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • the panel of genes consists essentially of the genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl , Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • the panel of genes consists of the genes Cyp2c6vl , F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • a gene panel of the present invention may for example include any of genes Cyp2c6vl , F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • a panel of genes may include genes from varied pathways and having varied activities.
  • a panel of genes may in one embodiment comprise mammalian genes.
  • a panel of genes comprises rat genes.
  • a panel of genes comprises mouse genes.
  • a panel of genes comprises human genes.
  • a panel of genes comprises canine genes.
  • genes of different species may have different names while still having the same function or similar function. Further, genes of different species having the same or similar function may share sequence homology. Alternatively, genes having the same or similar function may not be homologous at the gene level but their gene products may share amino acid sequence identity over part or all of an expressed polypeptide.
  • genes used in the methods, assays, uses, and arrays of this invention comprise mammalian genes.
  • genes comprise human genes.
  • genes comprise rat genes.
  • genes comprise mouse genes.
  • genes comprise canine genes.
  • genes comprise dog genes.
  • genes comprise pig genes.
  • a gene of a panel of genes encodes a cytochrome P450 family member, for example mouse gene Cyp2c6vl, which encodes a P450 family member of family 2, subfamily c, polypeptide 37.
  • the gene product of Cyp2c6vl has a functional role in metabolizing arachidonic acid to produce 12-hydroxyeicosatetraenoic acid (HETE).
  • a gene of a panel of genes encodes a coagulation factor IIII, for example mouse gene F3, which encodes a polypeptide that initiates blood coagulation by forming a complex with circulating factor VII or Vila.
  • the [TF:VIIa] complex activates factors IX or X by specific limited protolysis.
  • TF plays a role in normal hemostasis by initiating the cell-surface assembly and propagation of the coagulation protease cascade.
  • a gene of a panel of genes encodes a Galectin-2 polypeptide, for example mouse gene Lgals2, which encodes a polypeptide that binds beta-galactoside.
  • a gene of a panel of genes encodes a Latent- transforming growth factor beta-binding protein 4, for example mouse gene Ltbp4.
  • Ltbp4 encodes a polypeptide that may be involved in the assembly, secretion and targeting of TGFB1 (transforming growth factor beta-1) to sites at which it is stored and/or activated.
  • the protein may also play critical roles in controlling and directing the activity of TGFB1, and may have a structural role in the extra cellular matrix (ECM).
  • ECM extra cellular matrix
  • a gene of a panel of genes encodes an alpha 2-anti-plasmin, for example mouse gene SerpinG, which encodes a serine protease inhibitor. The major targets of this inhibitor are plasmin and trypsin, but it is also thought to inactivate matriptase-3/TMPRSS7 and chymotrypsin.
  • a gene of a panel of genes encodes a Tetratricopeptide repeat protein 36, for example mouse gene Ttc36, which encodes a protein that is a member of the TTC36 family.
  • a gene of a panel of genes encodes an Agrn protein, for example mouse gene Agrn, which encodes a protein that is thought to have acetylcholine receptor regulator activity.
  • a gene of a panel of genes encodes a C4b-binding protein beta chain, for example rat gene C4bpb, which encodes a protein that controls the classical pathway of complement activation.
  • the C4bpb polypeptide binds as a cofactor to C3b/C4b inactivator (C3bINA), which then hydrolyzes the complement fragment C4b.
  • the protein also accelerates the degradation of the C4bC2a complex (C3 convertase) by dissociating the complement fragment C2a.
  • the C4bpb encoded protein interacts also with anticoagulant protein S and with serum amyloid P component.
  • a gene of a panel of genes encodes a complement component 6, for example mouse gene C6, which encodes a protein involved in immune response.
  • a gene of a panel of genes encodes a caspase, for example Caspase-6 encoded by mouse gene Casp6.
  • Casp6 encodes a protein involved in the activation cascade of caspases responsible for apoptosis execution. The protein cleaves poly(ADP-ribose) polymerase in vitro, as well as lamins. Overexpression promotes programmed cell death.
  • a gene of a panel of genes encodes a Caveolin-2 protein, for example mouse gene Cav2.
  • Cav2 encodes a protein that may act as a scaffolding protein within caveolar membranes. The protein interacts directly with G-protein alpha subunits and can functionally regulate their activity. It further acts as an accessory protein in conjunction with CAV1 in targeting to lipid rafts and driving caveolae formation.
  • the Ser-36 phosphorylated form has a role in modulating mitosis in endothelial cells. Positive regulator of cellular mitogenesis of the MAPK signaling pathway. Required for the insulin- stimulated nuclear translocation and activation of MAPK1 and STAT3, and the subsequent regulation of cell cycle progression.
  • a gene of a panel of genes encodes a Cystathionine beta-synthase protein, for example mouse gene Cbs, which encodes the only known pyridoxal phosphate-dependent enzyme that contains heme. It is an important regulator of hydrogen sulfide, especially in the brain, utilizing cysteine instead of serine to catalyze the formation of hydrogen sulfide. Hydrogen sulfide is a gastratransmitter with signaling and cytoprotective effects such as acting as a neuromodulator in the brain to protect neurons against hypoxic injury.
  • a gene of a panel of genes encodes a Charged multivesicular body protein 4c, for example mouse gene Chmp4c, that is a probable core component of the endosomal sorting required for transport complex III (ESCRT-III) which is involved in multivesicular bodies (MVBs) formation and sorting of endosomal cargo proteins into MVBs.
  • a gene of a panel of genes encodes a Death associated protein kinase 1 , for example mouse gene Dapkl.
  • the protein is a calcium/calmodulin-dependent serine/threonine kinase involved in multiple cellular signaling pathways that trigger cell survival, apoptosis, and autophagy.
  • the protein regulates both type I apoptotic and type II autophagic cell deaths signal, depending on the cellular setting.
  • a gene of a panel of genes encodes a Fibulin-1, for example mouse gene Fblnl, which encodes a protein incorporated into fibronectin-containing matrix fibers.
  • the protein may play a role in cell adhesion and migration along protein fibers within the extracellular matrix (ECM), and could be important for certain developmental processes and contribute to the supramolecular organization of ECM architecture, in particular to those of basement membranes.
  • a gene of a panel of genes encodes a guanine deaminase, for example mouse gene Gda, which encodes an enzyme that catalyzes the hydrolytic deamination of guanine, producing xanthine and ammonia.
  • a gene of a panel of genes encodes a Hyaluronan-binding protein 2, for example mouse gene Habp2, which encodes a protein that cleaves the alpha-chain at multiple sites and the beta-chain between 'Lys-53' and 'Lys-54' but not the gamma-chain of fibrinogen and therefore does not initiate the formation of the fibrin clot and does not cause the fibrinolysis directly.
  • a gene of a panel of genes encodes a Sphingomyelin phosphodiesterase 3, for example mouse gene Smpd3, which encodes an enzyme that catalyzes the hydrolysis of sphingomyelin to form ceramide and phosphocholine. The enzyme also hydrolyzes sphingosylphosphocholine. Overexpression enhances cell death, suggesting that it may be involved in apoptosis control.
  • a gene of a panel of genes encodes a Tryptophan 2,3-dioxygenase, for example mouse gene Tdo2, which encodes a protein that incorporates oxygen into the indole moiety of tryptophan. It has a broad specificity towards tryptamine and derivatives including D- and L-tryptophan, 5-hydroxytryptophan and serotonin.
  • a gene of a panel of genes encodes a vitronectin, for example mouse gene Vtn, which encodes a cell adhesion and spreading factor found in serum and tissues. Vitronectin interacts with glycosaminoglycans and proteoglycans.
  • a gene of a panel of genes is selected based on its expression following incubation of a cell culture system with known hepatotoxic and non-hepatotoxic compounds. Each possibility represents a separate embodiment of the present invention.
  • a gene of this invention may be a homologue or an orthologue having similar function and or sequence, as the genes described herein. It shall be understood that variants, mutants, parts or homologous gene sequences having the same function, are included in the scope of definition as well as protection. The degree of alteration between the original sequence and its derivatives is inevitably limited by the requirement of altered gene expression by toxins.
  • the homology at the gene level amounts to at least 85%. In another embodiment, the homology at the gene level amounts to at least 65%. Possible alterations of a gene comprise deletion, insertion, substitution, modification and addition of at least one nucleotide, or the fusion with another nucleic acid. Each possibility represents a separate embodiment of the present invention.
  • gene product denotes molecules that are formed from the substrate of said genes by biochemical, chemical or physical reactions, such as DNA synthesis, transcription, splicing, translation, fragmentation or methylation.
  • Preferred gene products of the invention are RNA, particularly mRNA and cRNA, cDNA and proteins.
  • identity of the gene product amino acid sequence amounts to at least 85%.
  • identity of the gene product amino acid sequence amounts to at least 65%.
  • identity of the gene product amino acid sequence resides in active regions of the protein.
  • a "cellular system” refers in one embodiment to be any in vitro or ex vivo cell culture comprising cells.
  • the cellular system can be selected from the group of single cells, primary culture cells, cell cultures, tissues, and organs.
  • the scope of the cellular system also comprises parts of such biological entities, for examples samples of tissues and organs, i.e., ex vivo explants. Methods of growing and culturing different types of cellular systems are well known and understood by those skilled in the art.
  • the cellular system is of the same origin as the organism for which the unknown compound may be intended.
  • a new drug intended for administration to a human is tested in a cellular system comprising human cells.
  • the cellular system is of a different origin from the organism for which the unknown compound may be intended.
  • a new drug intended for administration to a human is tested in a cellular system comprising non-human mammalian cells.
  • Mammalian cells may in one embodiment be rat cells.
  • In a further embodiment may be monkey cells.
  • Each possibility represents a separate embodiment of the present invention.
  • the cells may be taken in vivo or in situ from a mammal.
  • the withdrawal of the cellular sample follows good medical practice.
  • a biological sample comprising cells and or tissue may be taken from any kind of biological species, but in certain embodiments of this invention, the samples is taken from a human, rat, mouse, pig, canine, or dog. Each possibility represents a separate embodiment of the present invention.
  • a "culture” refers in one embodiment, to any type of primary cells or genetically engineered cells or tissue culture cells or an ex vivo tissue explant, or a cell line, provided that they are capable of expressing at least the genes Cyp2c6vl, F3, Lgals2, Ltbp4, SerpinG, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl , Fblnl, Gda, Habp2, Smpd3, Tdo2 or Vtn
  • a "culture” comprises any type of primary cells or genetically engineered cells or tissue culture cells or an ex vivo tissue explant, or a cell line, provided that they are capable of expressing at least the genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • a "culture" gene capable of being expressed further comprise F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl , Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the cultures may originate from any tissue, including the uterus, pituitary gland, liver, brain, colon, breast, adipose tissue, etc.
  • a culture comprises liver tissue.
  • the culture comprises liver cells.
  • the culture comprises hepatocytes.
  • the culture comprises primary cells cultured to maintain physiological properties found in vivo.
  • physiological properties include cell morphology, cell-to-cell contact, appearance of cell membranes, formation of bile-canaliculi-like structures, or functionality of canaliculi-like structures, or any combination thereof.
  • the culture comprises primary culture cells cultured in the presence of extracellular matrix components, singly or as a mix, for example collagen or Matrigel Matrix (Corning), respectively.
  • extracellular matrix components for example collagen or Matrigel Matrix (Corning)
  • Methods of growing and culturing various cell cultures are well known by those skilled in the art.
  • Primary hepatocytes are the current gold standard for drug metabolism and CYP induction/inhibition studies in vitro.
  • primary culture mammalian hepatocytes are used in methods of this invention.
  • primary culture rat hepatocytes are used in methods of this invention.
  • primary culture mouse hepatocytes are used in methods of this invention.
  • primary culture canine hepatocytes are used in methods of this invention.
  • primary culture human hepatocytes are grown and maintained in a collagen sandwich culture.
  • cultures are maintained in serum-free medium.
  • cultures are maintained, in a complete medium.
  • Sandwich cultures may include additional elements, such as ECM components (natural or synthetic) in order to maintain hepatocytes in as near a physiological state as possible.
  • the hepatocyte culture may be cultivated for a certain period of time or may be immediately subjected to incubation with a compound to be tested for hepatotoxicity. Before incubating a culture with a compound to be screened, the cell culture is divided into multiple portions. In one embodiment, at least two portions are provided; one is used for screening while the other one serves as control. In another embodiment, the number of portions for screening exceeds the number of control portions. In some embodiments, numerous portions are subjected to a high-throughput screening.
  • Compounds for screening may, in one embodiment, include any compound, for example any chemical structure, drug, medicine, or food additive, being considered for administration to a human.
  • a compound is a natural product.
  • a compound is based on a natural product but is synthesized.
  • a compound is a synthetic compound or a combinatory chemistry derivative of a known compound, or compounds created in a series towards high-throughput target-binding screening.
  • a compound is a derivative, a metabolite, or a pro-drug.
  • a compound is a drug or medicine directed at treating a liver ailment.
  • a compound is a drug or medicine directed at treating a non-liver ailment.
  • compounds according to the invention can be used in their final non-salt form.
  • the present invention also encompasses the use of these compounds in the form of their pharmaceutically acceptable salts, which can be derived from various organic and inorganic acids and bases by procedures known in the art.
  • pharmaceutically acceptable salt and “physiologically acceptable salt”, which are used interchangeable herein, in the present connection refer in one embodiment to an active ingredient which comprises a compound according to the invention in the form of one of its salts, in particular if this salt form imparts improved pharmacokinetic properties on the active ingredient compared with the free form of the active ingredient or any other salt form of the active ingredient used earlier.
  • the compounds to be screened in the inventive method are not restricted in anyway.
  • the compounds are selected from the group of nucleic acids, peptides, carbohydrates, polymers, small molecules having a molecular weight between 50 and 1.000 Da and proteins. These compounds are often available in libraries. It is preferred to incubate a single compound within a distinct portion of the cell culture. However, it is also possible to investigate the cooperative effect of compounds by incubating at least two compounds within one portion. A further portion of cells is simultaneously incubated in the absence of the compounds, and functions as a control culture for comparison purposes.
  • screening comprises screening of a single compound.
  • screening comprises screening a single compound at multiple concentrations or dosages or forms thereof, of any combination thereof.
  • screening comprising screening a combination of compounds for hepatotoxicity activity.
  • each compound may be at a different concentration or dosage or in a different form.
  • a combination of two compounds comprises a multitude of combinations wherein each compound may be at a different concentration or dosage or in a different form.
  • a combination of compounds may include at least 2, 3, 4, 5, 6, 7, 8, 9, 10 or more compounds. Each possibility represents an embodiment of this invention.
  • the term "incubation” refers in one embodiment, to the contacting of the compound or compounds with the cells for a distinct period, which depends on the kind of compounds and/or target.
  • the incubation process also depends on various other parameters, e.g. the cell type and the sensitivity of detection, which optimization follows routine procedures known to those skilled in the art. Adding chemical solutions and/or applying physical procedures, e.g. impact of heat, can improve the accessibility of the target structures in the sample. Specific incubation products are formed as result of the incubation.
  • the duration of incubation may also be dependent on the state of the cells in culture, for example, ideally, cultures remain healthy, metabolically active and in as near a physiological representation to in vivo status as possible.
  • incubation is for at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 , 12, 13, 14, or 15 days. In another embodiment, incubation is for about 7 days. In another embodiment, incubation is for at least 7 days. In another embodiment, incubation is for about 14 days. In another embodiment, incubation is for at least 14 days. In another embodiment, incubation is for more than 14 days. In another embodiment, the results of 7 and 14 day incubations predict about the same or nearly the same level of hepatotoxicity or lack thereof. [0059] The incubation concentrations of compounds to be screens may be varied to best assess the hepatotoxicity of a compound.
  • the concentration of a compound being screened is in the range of an expected therapeutically effective amount.
  • expected therapeutically effective amount refers in one embodiment to an amount which, compared with a corresponding subject who has not received this amount, would be expected to have the following consequence: improved treatment, healing, prevention or elimination of a disease, syndrome, condition, complaint, disorder or side-effects or also the reduction in the advance of a disease, complaint or disorder.
  • the concentration used for incubation is twice the expected therapeutically effective amount.
  • the concentration used for incubation is up to 10-fold greater than the expected therapeutically effective amount.
  • the concentration used for incubation is 1 ⁇ 2 the expected therapeutically effective amount.
  • the concentration used for incubation is less-than 10-fold of the expected therapeutically effective amount.
  • expected therapeutically effective amount also encompasses the amounts which are expected to be effective for increasing normal physiological function. Testing of several compounds makes the selection of that compound possible that is best suited for the treatment of the mammal subject.
  • the identification of hepatotoxicity of compounds is indirectly performed in one embodiment, by determining the expression pattern of genes Cyp2c6vl, F3, Lgals2, Ltbp4, SerpinG, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 or Vtn, which the culture system is capable of expressing.
  • the expression pattern is of genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn, which the culture system is capable of expressing.
  • the expression pattern further includes expression of F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof, each of which the culture system is capable of expressing in any combination thereof.
  • the determination is performed at a specified moment and correlated to the signal strength at the beginning of the experiment and the positive/negative control. Either the control system is not incubated with the compounds (negative control) or the control system is incubated with a standard compound having no hepatotoxic activity (negative control) or a standard compound having hepatotoxic activity (positive control) as set forth in the Examples below. The hepatotoxic activity is revealed by a change in expression as compared with control cultures.
  • predicting hepatotoxicity takes into account all 20 genes within the gene panel (Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2 Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn).
  • predicting hepatotoxicity takes into account a subset of the 20 genes within the gene panel (Cyp2c6vl , F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn).
  • the subset is the 8 genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • predicting hepatotoxicity takes into account the 8 genes within the gene panel (Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn).
  • predicting hepatotoxicity further takes into account genes F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • a differential gene expression value is used to compute a score, wherein said score is correlated with a level of liver toxicity of a known compound, thereby predicting the liver toxicity of a compound.
  • the genes expressed or repressed in culture system incubated with compound having unknown hepatotoxicity are compared to the genes expressed or repressed in cells that were not exposed to the same compounds at the same concentrations. Pairwise comparisons are made between each of the treatments. A pairwise comparison is the expression data for a given gene under a given treatment condition compared to the expression data for this gene under a second treatment condition. The comparison is performed using suitable statistical technique with the assistance of known and commercially available programs.
  • each gene is given a weighted parameter based on the expression level of that gene, in a method or use of this invention for prediction hepatotoxicity values or scores.
  • a computational model is applied to an entire gene expression data set in order to produce a predictor value or score.
  • genes selected for a panel of genes for predicting hepatotoxicity are selected based on the correlation of their expression levels, weighted parameter values, or scores, or any combination thereof.
  • genes selected for a panel of genes for predicating hepatotoxicity are selected based on the correlation of their expression levels with known hepatotoxic and non-hepatotoxic compounds, weighted parameter values thereof, or scores thereof, or any combination thereof.
  • hepatotoxicity is predicted if the expression of genes is up- regulated or down-regulated in the culture system in comparison with a negative or positive control culture system, or if the expression of genes is substantially identical in the culture system and a positive control culture system.
  • the hepatotoxic activity is detected following the incubation step by differential gene expression analysis with the negative control system. Suitable tests for monitoring gene expression, determination and variant analysis of nucleotide sequences are known to those skilled in the art or can be easily designed as a matter of routine.
  • the assay according to the invention may be any assay suitable to detect and/or quantify gene expression.
  • the selected markers can be used to establish screening tools with a higher throughput, for instance, High Content Imaging (HCI) or a gene expression panel (e.g. real-time PCR-based TaqManTM low density arrays (TLDA) or bDNA assays on Luminex, nCount from Nanostring). Both technologies allow the combination of several selected endpoints, preserving biological complexity and molecular interactions to a certain extent. Especially HCI offers the possibility to combine classical hepatotoxic endpoints (e.g. micronuclei induction) and the analysis of cellular markers with the simultaneous acquisition of cell viability/ cytotoxicity. Cell viability is an important parameter to consider for hepatotoxicity testing because false positives in standard assays can be generated among others via cytotoxicity. Consideration of cytotoxicity for dose selection, together with multiple endpoint measurements may prevent or reduce false positives.
  • HCI High Content Imaging
  • TLDA real-time PCR-based TaqManTM low density arrays
  • hepatotoxic gene expression comparisons include the use of non- hepatotoxic compounds as controls.
  • hepatotoxic gene expression comparisons include using non-hepatotoxic compounds comprising at least one of entacapone, pioglitazone, N-acetyl-meta-aminophenol (AMAP), methanol, metformin, benzoic acid spectinomycin, streptomycin, theophylline, DL-lactic acid, metronidazole, penicillin V, progesterone, fiavoxate, quetiapine, minoxidil, dihydroergotamine, dimethyl sulfoxide (DMSO) or busipirone, or any combination thereof.
  • AMAP N-acetyl-meta-aminophenol
  • methanol metformin
  • benzoic acid spectinomycin streptomycin
  • streptomycin theophylline
  • DL-lactic acid metronidazole
  • penicillin V progesterone
  • hepatotoxic gene expression comparisons include the use of known hepatotoxic compounds as controls.
  • hepatotoxic gene expression comparisons include using known hepatotoxic compounds comprising at least one of azathioprine, carbamazepine, chloramphenicol, clofibrate, erythromycin, fiutamide, halperidon, imipramine, indomethacin, detoconazole, labetalol, methyldopa, propylthiouracil, terbinafine, tetracycline, tolbutamide, nitrofurantoin, phenytoin, clozapine, diclofenac, olanzapine, dexamethasone, captopril, furosemide, meloxicam, amiodarone, chlorpromazine, cyclosporine A, acetaminophen, EMD 335823 (l-(2-trifluoromethoxypheny
  • assays Many different types are known, examples of which are set forth below, including analyses by nucleotide arrays and nucleotide filters.
  • the hybridization conditions (temperature, time, and concentrations) are adjusted according to procedures also well known in the art.
  • methods of this invention apply chip hybridization and/or PCR for the determination of gene expression.
  • the assay of the invention involves the use of a high density oligonucleotide array.
  • the analysis is performed by multiplex qPCR, while in yet another embodiment, low density TaqMan arrays or branched DNA assays and used for analysis.
  • Other solid supports and microarrays are known and commercially available to the skilled artisan.
  • this invention relates to a method for predicting the hepatotoxicity of a compound by preparing a nucleic acid sample from a cell to be evaluated, contacting the nucleic acid sample with an microarray, detecting a nucleic acid hybridizing with the microarray, and comparing a result detected in steps quantitatively measuring gene expression, with a result detected using a nucleic acid sample prepared from a control cell, wherein hepatotoxicity may then be predicted.
  • RNA, cRNA, cDNA and/or protein are detected as the gene products, more preferably mRNA, cRNA and/or cDNA.
  • the total RNA from such cells is prepared by methods known to the skilled artisan such as by Trizol (Invitrogen) followed by subsequent re-purification, e.g. via Rneasy columns (Qiagen).
  • gene expression may be measured by quantitative analysis of the mRNA, RNA, cRNA, cDNA, and/or protein of a population of cells, for example a hepatocyte cell culture system. Each possibility represents another embodiment of this invention.
  • the total RNA is used to generate a labeled target according to methods and using detectable labels well-known in the art.
  • the RNA may be labeled with biotin to form a cRNA target for use in an assay.
  • cDNAs are produced using a reverse transcriptase (for example, Superscript Reverse Transcriptase; ThermoFisher) and labeled dNTP (for example, Cy3-dUTP and Cy5-dUTP; GE Healthcare Life Sciences), and a cDNA sample that reflects the amount of genes expressed within the cells to be evaluated is prepared. This causes labeled cDNA to be included in the cDNA sample.
  • a reverse transcriptase for example, Superscript Reverse Transcriptase; ThermoFisher
  • labeled dNTP for example, Cy3-dUTP and Cy5-dUTP; GE Healthcare Life Sciences
  • cDNA sample prepared in this manner is applied to the below-mentioned microarray in its single stranded denatured form, and cDNAs included in the cDNA sample are hybridized with the genes immobilized on the basal plate.
  • in situ hybridization is a methodology for determining the presence of or the copy number of a gene in a sample, for example, fluorescence in situ hybridization (FISH).
  • FISH fluorescence in situ hybridization
  • in situ hybridization comprises the following major steps: (1) fixation of tissue or biological structure to be analyzed; (2) pre-hybridization treatment of the biological structure to increase accessibility of target nucleic acid, and to reduce non-specific binding; (3) hybridization of the mixture of nucleic acids to the nucleic acid in the biological structure or tissue; (4) post- hybridization washes to remove nucleic acid fragments not bound in the hybridization; and (5) detection of the hybridized nucleic acid fragments.
  • probes used in such applications are typically labeled, for example, with radioisotopes or fluorescent reporters.
  • Preferred probes are sufficiently long, for example, from about 50, 100 or 200 nucleotides (nt) to about 1000 or more nucleotides, to enable specific hybridization with the target nucleic acid(s) under stringent conditions.
  • hybridization with cDNA can be accomplished, preferably by incubating at 50 to 80°C for 10 to 20 hours, more preferably about 65°C for 10 to 20 hours.
  • microarray refers in one embodiment, to nucleotide arrays that can be used to detect biomolecules, for instance to measure gene expression.
  • Array "slide” and “(DNA) chip” are used interchangeably in this disclosure having all the same meanings and qualities.
  • a microarray usually comprises a basal plate, e.g. made of slide glass, silicone, or the like, and DNA fragments immobilized as an array on this basal plate. With this microarray, DNAs contained in a sample can be detected by hybridizing them with the DNA fragments immobilized on the basal plate. Since the DNA within the sample is radiolabeled or fluorescently labeled, detection with radio imaging scanner, fluorescence imaging scanner, or the like is possible.
  • oligonucleotide arrays are made in research and manufacturing facilities worldwide, some of which are available commercially.
  • One of the most widely used oligonucleotide arrays is GeneChip made by Affymetrix, Inc.
  • the oligonucleotide probes have a length of 10 to 50 nucleotides (nt), preferably 15 to 30 nt, more preferably 20 to 25 nt. They are synthesized in-silico on the array substrate. These arrays tend to achieve high densities, e.g. more than 40,000 genes per cm2.
  • the spotted arrays tend to have lower densities, but the probes, typically partial cDNA molecules, usually are much longer than 25 nucleotides.
  • a representative type of spotted cDNA array is LifeArray made by Incyte Genomics. Pre- synthesized and amplified cDNA sequences are attached to the substrate of these kinds of arrays.
  • the array is a matrix, in which each position represents a discrete binding site for a product encoded by a gene, e.g. a protein or RNA, and in which binding sites are present for products of all genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • the array is a matrix, in which each position represents a discrete binding site for a product encoded by a gene, e.g.
  • binding sites are present for products of at least all genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • binding sites are present for further products of genes F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the "binding site” (hereinafter “site") is a nucleic acid or nucleic acid analogue to which a particular cognate cDNA can specifically hybridize.
  • the nucleic acid or analogue of the binding site can be, e.g. a synthetic oligomer, a full-length cDNA, a less-than full length cDNA or a gene fragment.
  • the microarray has binding sites for genes relevant to the action of the gene expression modulating agent of interest or in a biological pathway of interest.
  • more than one DNA fragment which is capable of hybridizing under stringent conditions to a gene or parts thereof as selected from the defined group Cyp2c6vl , F3, Lgals2, Ltbp4, SerpinG, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn, is immobilized on the basal plate.
  • more than one DNA fragment which is capable of hybridizing under stringent conditions to a gene or parts thereof as selected from the defined group Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn, is immobilized on the basal plate.
  • the defined group further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the DNA fragment to be immobilized on the basal plate may contain the whole or a part of the genes.
  • parts of a gene used herein means a portion of the gene and a nucleotide sequence equivalent to at least 10 nt, preferably at least 25 nt, more preferably 50 nt, most preferably 300 nt, highly preferably 500 nt.
  • genes constitutively expressing regardless of the presence or absence of chemical substances having hepatotoxic activity are immobilized on the basal plates of the microarray.
  • the expression level of the genes according to the invention can be corrected by immobilizing negative control genes on the basal plate and correcting the expression level of the negative control genes to a constant value.
  • the changes in the expression level of genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl , Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn can be detected with certainty.
  • the changes of expression level of genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn can be detected with certainty.
  • the nucleic acid or analogue are attached to a solid support or basal plate, which terms are used interchangeably herein, and which may be made from glass, plastic (e.g. polypropylene or nylon), polyacrylamide, nitrocellulose or other materials.
  • a solid support or basal plate which terms are used interchangeably herein, and which may be made from glass, plastic (e.g. polypropylene or nylon), polyacrylamide, nitrocellulose or other materials.
  • a conventionally known technique can be used.
  • the surface of the basal plate can be treated with polycations such as polylysines to electrostatically bind the DNA fragments through their charges on the surface of the basal plate.
  • techniques to covalently bind the 5'-end of the DNA fragments to the basal plate may be used.
  • a basal plate having linkers on its surface can be produced, and functional groups that can form covalent bonds with the linkers are introduced at the end of the DNA fragments.
  • the DNA fragments are immobilized by forming a covalent bond between the linker and the functional group.
  • a method for attaching the nucleic acids to a surface is by printing on glass plates.
  • cDNAs that hybridized with the DNA fragments on the microarray are detected.
  • the fluorescence is detected with, for example, a fluorescence laser microscope and a CCD camera, and the fluorescence intensity is analyzed with a computer.
  • detection can be carried out using an Rl image scanner and such, and the intensity of the radiation can be analyzed with a computer.
  • the detection of hepatotoxic activity can be additionally refined at the step measuring gene expression.
  • the gene expression is determined by detecting a respective gene product encoded by all or a subset of the genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl , Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn, and correlating an amount of signal or change in signal with the gene expression in the system.
  • a subset of the genes comprises Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of the genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the cellular culture system of the invention is incubated with various concentrations of a known hepatotoxic compound.
  • the amount of emitted signal or change in signal observed in the presence of the hepatotoxic compound is indicative of the change in gene expression experienced by the compound.
  • the change can be then related to the concentration of the hepatotoxic compound in the sample, i.e. the calibration curve enables the meter-reading of a matching concentration.
  • the calibration curve is based on the Lambert-Beer equation if using UV/VIS coloring or luminescence.
  • the differential expression of all 20 genes of a panel of genes for example genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn, is used by the classifier in a non-trivial way to predict toxicity.
  • the differential expression of a subset of all 20 genes of a panel of genes is used by the classifier in a non-trivial way to predict toxicity.
  • the subset further includes F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl , Fblnl , Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the classifier was used to get a toxicity score and to predict toxicity.
  • training and classification are two basic steps to using a classifier: training and classification.
  • Training is the process of taking content that is known to belong to specified classes, for examples gene expression based on exposure to known toxic verses non-toxic compound, and creating a classifier on the basis of that known content.
  • Classification is the process of taking a classifier built with such a training content set and running it on unknown content, for example gene expression in response to exposure to an unknown compound, to determine class membership for the unknown content.
  • Training is an iterative process whereby the best classifier possible is build, and classification is a one-time process designed to run on unknown content.
  • predicting hepatotoxicity comprises examining the differential expression of a panel of genes expressed by a hepatocyte culture system, wherein said culture is contacted with at least an unknown compound, different doses of an unknown compound, a combination of unknown compounds, or a combination of known and unknown compounds.
  • controls may also include incubation with known non-hepatotoxic compounds.
  • known concentrations are statistically proven, therefore representing a certain level or range, respectively.
  • the direction and strength of gene expression have also been figured out by the differential expression analysis of the target genes of the invention such that either a distinct up-regulation or down-regulation with a certain factor has been recognized as set forth below, which forms the basis of biomarker selection. Any measured concentration, which differs from the gene product concentration level of non-stimulated cells, indicates an abnormality of the tested cell sample, whereas a compound cannot be classified as hepatotoxic at a gene product concentration which is comparable to the concentration level of non- stimulated cells.
  • methods of this invention measure concentrations, which are higher than the gene product concentration level of non-stimulated cells, for detecting hepatotoxicity. Using this method, the inventors demonstrated sensitivity to micro molar ( ⁇ ) concentrations.
  • the "Polymerase Chain Reaction” or "PCR” is an amplification-based assay used to measure the copy number of the gene.
  • the corresponding nucleic acid sequences act as a template in an amplification reaction.
  • the amount of amplification product will be proportional to the amount of template in the original sample. Comparison to appropriate controls provides a measure of the copy number of the gene, corresponding to the specific probe used, according to the principle discussed above.
  • RNA RNA RNA RNA RNA RNA RNA RNA RNA RNA RNA RNA RNA RNA RNA transcribed from a given gene that is present in a cell or a biological sample.
  • a biological state of a biological sample e.g. a cell or cell culture
  • the transcriptional state of a biological sample includes the identities and abundances of the constituent RNA species, especially mRNAs, in the cell under a given set of conditions.
  • a substantial fraction of all constituent RNA species in the biological sample are measured, but at least a sufficient fraction is measured to characterize the action of a compound of interest.
  • the primers are designed based on the nucleotide sequence information of the region flanking the site to be amplified.
  • the primers may be designed so as to amplify a region of 100 to 200 base pairs in length.
  • the nucleic acid amplification method includes, but is not particularly limited to, a PCR, preferably a real-time PCR.
  • the level of mRNA may also be quantified by other methods described herein.
  • a primer may be labeled in advance.
  • fluorescent labels include FAMTM, TETTM, HEXTM, TAMRATM and ROXTM manufactured by Applied Biosystems.
  • either the 5'-end or the 3'-end of a primer may be labeled, preferably the 5'-end.
  • the nucleic acid may be labeled during PCR by using labeled nucleotides, or even after PCR is completed. Light emission is measured by a general-purpose luminescence determination device.
  • TaqMan-based assays use a fluorogenic oligonucleotide probe that contains a 5 -fluorescent dye and a 3'- quenching agent. The probe hybridizes to a PCR product, but cannot itself be extended due to a blocking agent at the 3'-end.
  • the 5'- nuclease activity of the polymerase for example, AmpliTaq, results in the cleavage of the TaqMan probe. This cleavage separates the 5 '-fluorescent dye and the 3-quenching agent, thereby resulting in an increase in fluorescence as a function of amplification.
  • the presence or absence of an amplified nucleic acid fragment can also be checked by subjecting a reaction solution to electrophoresis, such as for single-strand conformation polymorphism (SSCP) analysis, which may be performed by capillary electrophoresis.
  • electrophoresis such as for single-strand conformation polymorphism (SSCP) analysis, which may be performed by capillary electrophoresis.
  • SSCP single-strand conformation polymorphism
  • gel electrophoresis are also applicable and well known to those skilled in the art.
  • the present invention relates to the assessment or measurement of modulations of gene expression by the assays as set forth above.
  • modulation refers to the induction or inhibition of expression of a gene.
  • modulation of gene expression may be caused by endogenous or exogenous factors or agents.
  • the effect of a given compound can be measured by any means known to those skilled in the art. For example, expression levels may be measured by PCR, Northern blotting, Primer Extension, Differential Display techniques, etc.
  • the induction of expression i.e. up-regulation refers to any observable or measurable increase in the levels of expression of a particular gene, either qualitatively or quantitatively. Contrary to that, the inhibition of expression (i.e.
  • down-regulation refers to any observable or measurable decrease in the levels of expression of a particular gene, either qualitatively or quantitatively.
  • the measurement of levels of expression may be carried out using any techniques that are capable of measuring RNA transcripts in a biological sample. Examples of these techniques include, as discussed above, PCR, TaqMan, Primer Extension, Differential display and nucleotide arrays, among other things.
  • Ct cycle threshold or detectable cycle
  • Ct levels are inversely proportional to the amount of target nucleic acid in the sample (ie the lower the Ct level the greater the amount of target nucleic acid in the sample).
  • the earlier a fluorescent signal passes the threshold value the higher is the expression of that gene in that sample.
  • the inverse relationship of Ct levels to the amount of target nucleic acid differs from an Illumina microarray platform, where there is a positive correlation between the expression level and the signal. Thus, we expect a negative correlation between the expression values in the qPCR and the Illumina microarray platforms.
  • Ct parameter provides a numerical value, also known as a score that is inversely proportional to the abundance of a verified cDNA species in the original amplified template.
  • Ct values used in the methods, processes, and uses of this invention are normalized.
  • predicting hepatotoxicity comprises using normalized gene expression values as input for a classifier and obtaining a toxicity score. The score may then be compared to a threshold to determine toxicity.
  • the panel of marker genes of the invention exhibits a sensitivity that allows in one embodiment the use of only 6 marker genes in the scope of the screening method
  • methods, uses and arrays of this invention comprise applying more than these marker genes for detecting hepatotoxicity.
  • the use of 6-20 marker genes provides the sensitivity.
  • the use of 2 marker genes provides the sensitivity.
  • Example 6 below shows that high classification rates are obtained by using 20 genes.
  • the hepatocyte culture system provided is capable of expressing at least the 20 genes of Table 3. Accordingly, in another embodiment, the expression of the 20 genes of Table 3 is compared with the gene expression in the control culture system following the incubation step.
  • the gene expression pattern of multiple genes of Table 3 may be measured and compared with the expression pattern in the control culture system, the hepatotoxicity can be characterized compound-specifically.
  • the expression pattern is determined by a correlation of the multiple genes and/or a magnitude of altered regulation.
  • the screening method of this invention not only evaluates the effect of chemical substances having hepatotoxic activity on cells to be evaluated, but can also indicate the details of this effect.
  • genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl , Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn may provide prediction to possible different mechanism of hepatotoxicity, not just a trivial hepatotoxic or not hepatotoxic answer.
  • a subset of genes from a gene panel of genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn may provide prediction to possible different mechanism of hepatotoxicity, not just a trivial hepatotoxic or not hepatotoxic answer.
  • a subset of genes comprises genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • additional genes includes with a gene panel of genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn may provide prediction to possible different mechanism of hepatotoxicity, not just a trivial hepatotoxic or not hepatotoxic answer.
  • Figure 8 schematically illustrates a flowchart of a method of predicting hepatotoxicity of a compound.
  • hepatocytes are isolated.
  • hepatocytes of rat primary hepatocytes In one embodiment, viability testing is performed using a Trypan blue staining.
  • QA1 ; 8003 hepatocyte cultures isolated and having a cell viability of less than 84% are discarded, whereas isolations with at least 84% viability proceed to the next step of establishing a hepatocytes culture system, for example a hepatocyte sandwich culture, and incubating the hepatocyte culture system with to a "test" compound (8005) to be screened.
  • the term "incubating” and “exposing” may be used interchangeably having all the same meanings and qualities. Further, the hepatocytes culture is dividing into at least two sub-cultures prior to the incubation step (8005) in order to provide control cultures and the ability to screen multiple compounds at the same time.
  • hepatocytes cultured for example in a collagen sandwich culture, are further analyzed to ensure the high quality of the culture, wherein functional and morphological properties are assessed.
  • Quality Assurance step 2 QA2; 8007 only cells that maintain the features and functionality required as described, for example in Example 2, proceed to step 8009, where nucleic acids are prepared from the hepatocyte culture system.
  • Nucleic acids used in methods of this invention may for example include RNA and cDNA.
  • RNA integrity is essential to ensure that high-quality starting material is used for the gene expression measurements. Isolation of RNA may use any methods known in the art as described above and in the Examples. The total RNA isolated, may for example be analyzed using the RNA Screen Tape 2200 TapeStation instrument and software (Agilent Technologies, CA, USA). The integrity of the total RNA (RIN) is determined and only RNA samples with a RIN above 8 are used for quantitative Polymerase Chain Reaction on TaqMan-Low Density Arrays (qPCR TLDAs) (8013).
  • qPCR TLDAs quantitative Polymerase Chain Reaction on TaqMan-Low Density Arrays
  • the gene expression results are then analyzed using the classifier software (8015).
  • the final classifier receives as input a list of 20 normalized expression values (normalized Ct values) for the Biomarker Signature Genes.
  • normalized Ct values normalized Ct values
  • the classifier calculates a toxicity score between 0 and 1, scores above a threshold are used to predicted hepatotoxicity. The initial threshold score was set to be 0.5 required (8017) (See Example 5below).
  • gene expression results are compared with control hepatocyte cultures either incubated with known hepatotoxic or non-hepatotoxic compounds, or without any additions.
  • the identification of a plurality of genes provides a powerful tool for assessing the hepatotoxicity of a compound.
  • This invention further provides a process of computationally constructing a panel of genes for predicting hepatotoxicity of compounds, comprising the steps of: (a) providing a hepatocyte culture expressing a physiological range of genes; (b) measuring the expression level of the genes from the hepatocyte culture; (c) removing from consideration all genes from step (a) with an expression level below a set threshold; (d) incubating at least a portion of the hepatocyte culture system with either a known hepatotoxic compound or a known non- hepatotoxic compound, wherein multiple incubating steps are performed each with at least a portion of a hepatocyte culture system, and each with either a known hepatotoxic compound or a known non-hepatotoxic compound; (e) measuring the gene expression of said remaining genes following said incubation step (d); (f) training a classifier based on gene expression
  • a process of this invention provides a panel of 60 genes for predicting hepatotoxicity of compounds.
  • a subset of the resultant panel of genes may be used for predicting hepatotoxic compounds.
  • about 20 genes of the 60 genes used may be used for predicting hepatotoxicity.
  • incubation steps of a process of this invention comprise the use of control known non-toxic compounds comprising entacapone, pioglitazone, N-acetyl-meta- aminophenol (AMAP), methanol, metformin, benzoic acid spectinomycin, streptomycin, theophylline, DL-lactic acid, metronidazole, penicillin V, progesterone, flavoxate, quetiapine, minoxidil, dihydroergotamine, dimethyl sulfoxide (DMSO) or busipirone, or any combination thereof.
  • control known non-toxic compounds comprising entacapone, pioglitazone, N-acetyl-meta- aminophenol (AMAP), methanol, metformin, benzoic acid spectinomycin, streptomycin, theophylline, DL-lactic acid, metronidazole, penicillin V, progesterone, flavoxate, quetiapine, minoxidil, dihydr
  • incubation steps of a process of this invention comprise the use of control known toxic compounds comprising azathioprine, carbamazepine, chloramphenicol, clofibrate, erythromycin, fiutamide, halperidon, imipramine, indomethacin, detoconazole, labetalol, methyldopa, propylthiouracil, terbinafme, tetracycline, tolbutamide, nitrofurantoin, phenytoin, clozapine, diclofenac, olanzapine, dexamethasone, captopril, furosemide, meloxicam, amiodarone, chlo ⁇ romazine, cyclosporine A, acetaminophen, EMD 335823, fenofibrate, rosiglitazone, troglitazone, or valproic acid, or any combination thereof.
  • control known toxic compounds comprising
  • high-dose comprises a dose above an expected therapeutically effective amount. In another embodiment, a high-dose is at least twice the expected therapeutically effective amount. In another embodiment, a high-dose is at least three-times the expected therapeutically effective amount. In another embodiment, a high-dose is at least four- times the expected therapeutically effective amount. In another embodiment, a high-dose is at least ten- times the expected therapeutically effective amount.
  • low-dose comprises a dose at an expected therapeutically effective amount. In another embodiment, a low-dose is at least half the expected therapeutically effective amount. In another embodiment, a low-dose is at least a third the expected therapeutically effective amount. In another embodiment, a low-dose is at least a tenth the expected therapeutically effective amount.
  • This invention further provides in one embodiment, a use of a panel of genes comprising genes Cyp2c6vl , F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn as marker genes for screening a compound for hepatotoxic activity.
  • a subset of the panel of genes comprises the marker genes.
  • the subset of the panel of genes comprises Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of the panel of genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl , Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • this invention provides use of nucleic acid probes specifically hybridizing under stringent conditions with a panel of genes comprising genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn, or gene products encoded by said panel of genes, or respective parts thereof, for detecting and quantitating the expression of said panel of genes, wherein said panel of genes is representative for a hepatotoxic activity in a cellular system.
  • nucleic acid probes specifically hybridizing under stringent conditions with a panel of genes comprising a subset of genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn, or gene products encoded by said panel of genes, or respective parts thereof, for detecting and quantitating the expression of said panel of genes, wherein said panel of genes is representative for a hepatotoxic activity in a cellular system.
  • the subset of genes comprises Cyp2c6vl , Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • Methods and uses of this invention comprise in certain embodiments the use of primers for specifically amplifying a region, including genes or portions thereof of member of a gene panel as provided herein, while a probe, for example a TaqMan probed provides quantitative visualization of the amplified region. (See above for Ct values).
  • probes may create a visible signal only when hybridized to the amplified product.
  • hybridization may be to a native DNA, a native RNA, an mRNA, a cDNA or to an amplified RNA or DNA.
  • nucleic acid refers to a natural or synthetic polymer of single- or double- stranded DNA or RNA alternatively including synthetic, non-natural or modified nucleotides, which can be incorporated in DNA or RNA polymers. Each nucleotide consists of a sugar moiety, a phosphate moiety, and either a purine or pyrimidine residue.
  • the nucleic acids are preferably single or double stranded DNA or RNA, primers, antisense oligonucleotides, ribozymes, DNA enzymes, aptamers and/or siRNA, or parts thereof.
  • the nucleic acids can be optionally modified as phosphorothioate DNA, locked nucleic acid (LNA), peptide nucleic acid (PNA) or aptmer.
  • a “nucleic acid probe” is a nucleic acid capable of binding to a target nucleic acid or complementary sequence through one or more types of chemical bond, usually through complementary base pairing by hydrogen bond formation.
  • a probe may include natural (i.e. A, G, C, or T) or modified bases (e.g. 7-deazaguanosine, inosine, etc.).
  • the bases in a probe may be joined by a linkage other than a phosphodiester bond, so long as it does not interfere with hybridization. It will be understood by one of skill in the art that probes may bind target sequences that lack complete complementarity with the probe sequence depending upon the stringency of the hybridization conditions.
  • the probes are preferably directly labeled with isotopes, e.g. chromophores, luminphores or chromogens, or indirectly labeled with biotin to which a streptavidin complex may later bind.
  • isotopes e.g. chromophores, luminphores or chromogens
  • biotin to which a streptavidin complex may later bind.
  • nucleic acid probes to be used as hepatotoxicity-specific markers are oligonucleotide probes.
  • array probes may be used for hybridization, as primers for amplification, or for qPCR, or any combination thereof.
  • probes create a visible signal only when hybridized to the amplified product.
  • the hybridization may in certain embodiments be to native mRNA, to cDNA, or to amplified DNA, or any combination thereof.
  • the specific markers can be labeled, in doing so the labeling depends on their inherent features and the detection method to be applied.
  • the applied methods depend on the specific incubation products to be monitored and are well known to the skilled artisan. Examples of suitable detection methods according to the present invention are fluorescence, luminescence, VIS coloring, radioactive emission, electrochemical processes, magnetism or mass spectrometry.
  • labeling method is not particularly limited as long as a label is easily detected.
  • labeled nucleic acid or oligonucleotide probe refers in one embodiment to one that is bound, either covalently through a linker or a chemical bond, or noncovalently through ionic, van der Waals, electrostatic, hydrophobic interactions or hydrogen bonds, to a label such that the presence of the nucleic acid or probe may be detected by detecting the presence of the label bound to the nucleic acid or probe.
  • the nucleic acids are labeled with digoxigenin, biotin, chemiluminescence substances, fluorescence dyes, magnetic beads, metallic beads, colloidal particles, electron-dense reagents, enzymes; all of them are well-known in the art, or radioactive isotopes.
  • isotopes for labeling nucleic acids in the scope of the invention may include 3 H, 14 C, 32 P, 33 P, 35 S or 1251. In another embodiment, isotopes include 32 P, 33 P, or 125 I.
  • this invention relates to a gene array comprising a defined combination of gene pluralities according to any Table or Figure or Example herein.
  • this invention provides an array for screening a compound for hepatotoxic activity, comprising nucleic acid probes that are capable of specifically hybridizing under stringent conditions with an amplified panel of genes, said panel of genes comprising Cyp2c6vl, F3, Lgals2, Ltbp4, SerpinG, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn, or gene products encoded by said panel of genes, or respective parts thereof.
  • the panel of genes comprises a subset of genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • the subset of genes comprising a panel of genes comprises genes Cyp2c6vl , Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the TLDA of this invention comprise in one embodiment, arrays of amplification products. These amplification products comprise genes of interest, for example the genes included in a panel of genes described herein. .
  • the expressed genes from a hepatocyte culture system may then be hybridized with these TaqMan probes.
  • the expressed genes are amplified during the PCR cycles.
  • the amplification levels correlate with the expression levels of each specific gene in the sample, a hybridization step is performed, TaqMan probe to each specific gene, which is followed by an amplification of the specific gene from the culture system sample.
  • the arrays of the invention may include an article that comprises written instructions or directs the user to written instructions for how to practice the method of the invention.
  • the prior teaching of the present specification concerning the screening method is considered as valid and applicable without restrictions to the array and any kit comprising an array of this invention if expedient.
  • a method for screening compounds with hepatotoxic activity which applies unique gene expression patterns of genes selected from the group comprising Cyp2c6vl , F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn, is provided for the first time.
  • genes selected comprise Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • the genes selected comprise a subset of genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl , Gda, Habp2, Smpd3, Tdo2 and Vtn.
  • the subset comprises genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • the present invention provides characteristic expression fingerprints of a subset of marker genes that are associated with hepatotoxicity. Statistical data analysis revealed that a panel of 20 genes may be used as representative for the predicting hepatotoxicity response. The data provided in the Examples below support that mechanistic profiling in vitro is a powerful tool compared to single endpoint detections to predict hepatotoxicity.
  • the 20 marker genes identified can be used in the future for the characterization of the hepatotoxic potential of unknown compounds as described herein.
  • a subset of the 20 marker genes is identified that can be used in the future for the characterization of the hepatotoxic potential of unknown compounds as described herein.
  • the subset includes 8 marker genes.
  • the subset includes more than 8 marker genes.
  • the subset includes at least 8 marker genes. In still another embodiment, between 8 and 20 can be used.
  • the analysis of the differential expressed genes is particularly suitable for higher throughput test systems.
  • chemicals can be identified with an unknown mode of action and predicting their potential to exert hepatotoxic effects.
  • the detection method as well as arising monitoring method of the invention can be performed in a simple and fast manner.
  • the appropriate array is cost-efficiently produced.
  • the genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn are qualified as biomarkers for detecting and characterizing hepatotoxicity.
  • Targeting gene products encoded by said genes is highly specific for the hepatotoxic activity.
  • a subset of the genes Cyp2c6vl, F3, Lgals2, Ltbp4, Serpinf2, Ttc36, Agrn, C4bpb, C6, Casp6, Cav2, Cbs, Chmp4c, Dapkl, Fblnl, Gda, Habp2, Smpd3, Tdo2 and Vtn are qualified as biomarkers for detecting and characterizing hepatotoxicity.
  • the subset of genes comprises genes Cyp2c6vl, Ttc36, Agrn, C6, Casp6, Cav2, Chmp4c, and Vtn.
  • the subset of genes further comprises F3, Lgals2, Ltbp4, Serpinf2, C4bpb, Cbs, Dapkl, Fblnl, Gda, Habp2, Smpd3, or Tdo2, or any combination thereof.
  • Targeting gene products encoded by said genes is highly specific for the hepatotoxic activity. All substance probes are characterized by a high affinity, specificity and stability as well as low manufacturing costs and convenient handling. These features form the basis for a reproducible action, wherein the lack of cross-reactivity is included, and for a reliable and safe interaction with their matching target structures.
  • the term “about”, refers to a deviance of between 0.0001-5% from the indicated number or range of numbers. In one embodiment, the term “about”, refers to a deviance of between 1 -10% from the indicated number or range of numbers. In one embodiment, the term “about”, refers to a deviance of up to 25% from the indicated number or range of numbers.
  • the term “comprising” is intended to mean that the system includes the recited elements, but not excluding others which may be optional.
  • the phrase “consisting essentially of it is meant a method that includes the recited elements but exclude other elements that may have an essential significant effect on the performance of the method. "Consisting of shall thus mean excluding more than traces of other elements. Embodiments defined by each of these transition terms are within the scope of this invention.
  • FIG. 1A An example of the distribution for two Illumina bead chip arrays is presented in Figures 1A and IB. (Illumina Inc. San Diego, CA. product # Rat-ref 12. Catalog IDs: BD-27- 303, BD-27-302) The 'noise' level for both arrays was about 2.0.
  • Figure 1A the array described by the solid line has relatively more 'noise' in expression level than the dashed line array, and consequently its tail distribution (higher than 'noise') is slightly lower.
  • one sample was chosen to be a reference, and the genes were divided into subsets of size 101, i.e., 101 genes per subset, ordered by the expression values for this reference sample.[ For each sample and each gene subset, the medians of expression in the subset were calculated for the sample and the reference. The normalized expression values in a gene subset for a sample were obtained by adding a constant value to all genes in the subset in order to make the median equal to the reference median in this subset.
  • the sets of toxic and non-toxic samples had to be defined based on the Merck-Serono toxicity labels.
  • a sample consisted of one experiment, for example a hepatocyte culture system or a portion thereof, treated with a compound at a single dose. This was followed by measuring the gene expression for all genes from this culture system sample. Data was collected from several samples for each compound and dose thereof. The DMSO negative controls and both doses of the non-toxic labeled samples were considered as non-toxic.
  • the genes' contributions to hepatotoxicity prediction was measured by training a classifier on the expression data.
  • the classifier used was Random Forest, implemented in R with default parameters.
  • One of the results of training a Random Forest is the variable importance, which measures the contribution of each parameter to the resulting classifier.
  • For classifier training all of the samples in the training set were assigned toxicity labels based on the toxicity of the compounds in Table 1 as follows. Samples treated with non-toxic compounds (all doses) and negative controls were assigned the label 'non-toxic'. Samples treated with toxic compounds with the higher dose were assigned the label 'toxic'. The toxicity labels for the compounds are shown in Table 1. The low doses for toxic compounds presented a problem since it was unclear whether the compounds were still toxic at these lower concentrations.
  • a further computational filter was applied to the expression profiles of the genes to increase the robustness of the results.
  • the expression distributions for high dose toxic and non- toxic samples was compared for each gene, wherein the medians were required to be at least 0.1 apart and the distribution overlap was restricted. The reasoning was that the next expression measurement platform would be new, therefore a new source of noise. Keeping the hepatotoxic and non-hepatotoxic expression distributions as far apart as possible reduced the risk that some genes would perform badly in the new platform. After applying this filter the gene list was reduced to 97 genes ( Figure 7, 7007). [00143] The remaining genes went through a manual selection stage.
  • a rat primary hepatocytes model was used for further experiments. This model developed according to previous experiments performed at Merck Serono (Tuschl et al. (2009) Chem. Biol. Interact. 181(1):124-137). Based on this model, primary rat hepatocytes were isolated from male Wistar rats (weighing 200-300gr) using a modified two-step perfusion method (Seglan et al, (1976) Methods Mol. Biol. 13:29-83).
  • Hepatocytes with viability above 84% were cultured between two layers of collagen (a collagen sandwich culture) in a serum free media as described previously (Tuschl and Muller, (2005) Toxicology 218(2-3): 205-215; Tuschl et al, (2009) Ibid.).
  • the hepatocytes cultured in the serum free collagen sandwich culture retained specific hepatocyte properties closest to physiological levels for more than 14 days.
  • the functional and morphological properties of the cells were verified by performing several standard functional tests on the cells (Tuschl et al., (2009) Ibid). All of the standard tests that are described below for assessing the functional status and viability of the cells were performed regularly for each new primary hepatocytes isolation, at the beginning of production and over the growth period of the cells.
  • Carboxy-DCFDA is transported specifically by Mrp2 and is considered as a valid marker for canalicular-like transport (Tuschl et al, (2009) Ibid).
  • a perfect and stable co-localization of canaliculi-like structures and the sites of fluorescence signals was observed at least until day 16 after seeding (data not shown). This indicates direct and active transport, and accumulation of the fluorescent dye into the extracellular space.
  • P450 enzymes catalyze the first step (phase I) in xenobiotic metabolism which can lead to detoxification as well as the formation of hepatotoxic metabolites (bioactivation).
  • the metabolic activity of the primary hepatocytes was verified by measuring basal and induced activities of P450 enzymes: CYP1A, CYP2B, CYP2C, and CYP3A by fluorimetric (EROD- CYPIA, BROD-CYP2B) or luminometric (2C-Glo-CYP2C, 3A-Glo-CYP3A) measurements of resorufin and luciferin generation, respectively ( Figure 3B).
  • Induced activity was measured after treatment with prototypical inducers at days 4, 7, 10 and 14 after seeding and incubation of hepatocytes for 48h ( Figure 3B).
  • Cultures were treated with ⁇ beta-naphtoflavone (BNF) for induction of CYPIA, with 500 ⁇ phenobarbital (PB) for induction of CYP2C and CYP2B, and with 50 ⁇ dexamethasone (DEX) for induction of CYP3A.
  • BNF beta-naphtoflavone
  • PB phenobarbital
  • DEX dexamethasone
  • Albumin synthesis is a classical assay for the study of liver-specific function (Suzuki et al., (1993) Cytotechnology. 11 :213-218).
  • the amount of albumin secreted to the medium was quantified in the collagen sandwich primary hepatocytes using an enzyme-linked immunosorbent assay (ELIZA).
  • ELIZA enzyme-linked immunosorbent assay
  • Conditioned medium was collected from cell culture at day 3,7,9,13,16 and the amount of albumin was measured using the Albumin Rat Eliza Kit (Abeam).
  • Figure 3D shows levels of Albumin synthetized by rat primary hepatocytes in a collagen sandwich culture. The amount of albumin secreted from the cells was stable over the period of growth (16 days), suggesting for a functional hepatocytes.
  • qRT-PCR TaqMan Real Time PCR
  • TLDA TaqMan low density array
  • the ability of the qRT-PCR technology to reproduce the results from Illumina microarray technology was verified by using TaqMan primers for the selected 60 genes.
  • the TaqMan primers for the 60 genes were placed on a set of 3 TLDAs.
  • Each TLDA is a 384-well array spotted with the TaqMan probes for the specific genes of interest, and additional four endogenous control genes that are used for normalization. Eight samples were analyzed per card and each sample was measured in duplicates in a single RT-PCR run. PCR cycle numbers (CT values were obtained and further analyzed.
  • RNA samples were then isolated from a sandwich culture of rat primary hepatocytes, treated by high dose toxic compounds, non-toxic compounds (high and low dose), and with buffer only (0.5% DMSO) control samples (Table 1 above lists compounds and doses.).
  • RNA samples were then isolated from a sandwich culture of rat primary hepatocytes, treated by high dose toxic compounds, non-toxic compounds (high and low dose), and with buffer only (0.5% DMSO) control samples (Table 1 above lists compounds and doses.).
  • four housekeeping genes were added to the TLDAs and used for normalization: RplpO, B2m, PSmd7, and Ddx47. These genes were carefully selected, to ensure that their expression was constant and stable, and was not influenced by treatment with various compounds.
  • the measured gene expression values for each sample were normalized by subtracting the average expression of the housekeeping genes from each sample, i.e., from each experimental performed in one well of cultured hepatocytes that were incubated with a specific compound or that were not treated (control). This normalization was performed for the genes of each array separately.
  • Ct levels are inversely proportional to the amount of target nucleic acid in the sample (ie the lower the Ct level the greater the amount of target nucleic acid in the sample), as opposed to other microarray platforms, for example an Illumina microarray platform, show a positive correlation between the expression level and the signal. Thus, it is expected that there will be a reverse correlation between the expression values in the qPCR and the Illumina microarray platforms.
  • FIGS. 4A-4C show the correlation between the two different platforms (TLDA/ microarray). This correlation shows a correspondence between the platforms.
  • each TLDA card in the 3-card set included a different subset of three of these genes. It was necessary to choose the preferred subset of three housekeeping genes and to finalize the normalization procedure for further experiments. For each set of three housekeeping gene the expression values of all cards were normalized using the average of only those genes. The different normalized expression values obtained were very similar and the correlations to Illumina expression data changed very little.
  • the selected housekeeping genes for further experiments were B2m, PSmd7 and Ddx47, correlations of expression using this subset were slightly better than the other combinations.
  • a 'transition' experiment was performed in order to achieve several goals.
  • the first goal was to verify the reproducibility of the gene expression data in rat primary hepatocytes that were isolated and grown as described herein.
  • the second goal was to test the possibility of shortening the duration of treatment with the tested compound from 14 days to 7 days.
  • Rat primary hepatocytes were isolated and seeded in a collagen sandwich culture. Three days after seeding, the cells were treated with one of the test compounds, which was dissolved in DMSO 0.5% v/v .Cells treated with the buffer only (0.5%DMSO) were used as controls. During the cell growth period, the standard tests as described above were performed in order to assess the functionality and viability of the cells. The cells were treated daily with the test compounds or vehicle controls for 7 days or for 14 days. After 7 and 14 days of treatment (10 or 17 days from seeding), cells were harvested and total RNA was isolated from each treated sample. RNA was analyzed using the RNA Screen Tape 2200 TapeStation instrument and software (Agilent Technologies, CA, USA).
  • RNA samples with a RIN above 8 were used for next steps.
  • RNA was reverse transcribed to produce cDNA.
  • RNA samples used were considered to have good quality RNA.
  • the quality and quantity of the cDNA was determined using the standard assays.
  • cDNA was used for the measurement of gene expression levels using quantitative real time PCR on TLDA. The experiment was performed in 6 replicates for each treatment, and from at least 2 different cell isolations.
  • Example 4 Following the results of Example 4, the 7 day protocol for cell treatment was adopted in the following experiments. In order to create gene expression information for all 60 genes selected in Example 1 in the current list and to extend the set of hepatotoxic and non-hepatotoxic compounds, additional experiments were performed.
  • Table 2 lists the compounds that were tested on primary rat hepatocytes.
  • the basic step was to choose a subset of genes from the list of 60 genes, train a classifier only on the expression of these genes with toxicity labels as the target, and measure the performance of the resulting classifier.
  • the classifier used was Random Forest as implemented in R (http://www.R-project.org, Random Forest package). The performance was measured using a form of cross validation where the score for samples treated with a compound is obtained by training a classifier on all other samples and applying it to the samples not involved in training ('leave compound out'). This overcomes the problem of similarity of expression values within replicates, and simulates the application of a classifier to expression related to a new compound.
  • the final gene list was selected in several rounds.
  • Chmp4c chromatin modifying protein 4C Rn01461787_ml
  • F3 coagulation factor III (thromboplastin, tissue factor) Rn00564925_ml
  • Serpinf2 serine (or cysteine) peptidase inhibitor clade F, Rn01464595_ml
  • the final classifier was trained using the expression data for the 20 genes of the Biomarker Signature, using the parameters obtained in the optimization, and ignoring the samples treated by compounds that had consistent prediction failures.
  • the final classifier receives as input a list of 20 normalized expression values for the Biomarker Signature Genes. It then calculates a toxicity score between 0 and 1, scores above a threshold are predicted as toxic.
  • the initial threshold score was set to be 0.5.
  • the data was collected using a reproducible primary rat hepatocyte in vitro platform that closely mimics liver drug induced response and retained specific hepatocyte properties closest to physiological levels for more than 7 days, thus, enabling assessment of gene expression levels after prolonged treatment with the tested drug.
  • the system was designed to predict chronic hepatotoxicity potential. This technique enables early detection of potentially hepatotoxic compounds, which can be removed from the drug development process, leading to a safer drugs and a more efficient drug discovery and development process.
  • Example 6 Classifier Validation
  • Rat primary hepatocytes, in a collagen sandwich culture were treated for 7 days with the selected compounds. After 7 days treatment expression levels of the 20 genes that constitute the final classifier, i.e., the "Biomarker Signature Genes" of Example 5, were measured by qRT- PCR on TLDAs. The normalized expression values of a sample (after sample normalization and gene normalization) were used as input for the classifier. The toxicity score was obtained and used for performance evaluation. The performance of the classifier is presented in Table 4. Total correct prediction was 85% with 86% sensitivity and 83% specificity. These results validate and confirm developing and using the Classifier to identify hepatotoxic compounds as described here.
  • Biomarker Signature Genes may be used as a panel of marker genes for screening compounds in order to predict hepatotoxicity, which may be particularly useful for unknown compounds and/or for determining dosage of known compounds and unknown compounds.
  • a beta-test was performed with the potential lead compounds. Twenty (20) compounds, consisting of real drug candidates in development and several references compounds, were received (blinded) and were blind tested to evaluate their toxicity with the classifier. The test conclusions and predictions include for each compound the Median Toxicity Score, which is a numerical score between 0 to 1 that reflects the level of toxicity according to classifier prediction, in addition to a Toxicity Level grade that is defined in accordance with the toxicity score (Severe Toxic (ST)->0.9, Toxic (T)-0.65-0.9, Borderline-0.65-0.55,Non-Toxic-(NT)- ⁇ 0.55). Additional information regarding the putative mechanism of toxicity triggered by the tested compound was recorded.
  • the test also has the ability to cluster the text compounds into subgroups according to their expression behavior as is illustrated in Figure 10 and Table 5 below.
  • the samples were those wherein their expression profile was correlated with a label of toxicity. DMSO expression values were used for normalization and were not used as non-toxic control in the training process. The final list of compounds for training contained 33 toxic and 14 non-toxic compounds.
  • a training dataset was obtained by taking the expression values for the 20 genes in the signature (Example 5, Table 3 above), for the samples selected.
  • a second training dataset was created using the 'DMSO transformation procedure' on the first dataset as follows: expression values close to the DMSO median (up to 0.5 in normalized log2 expression) were changed to the exact DMSO median. The purpose of this transformation was to avoid classifier decisions based on values too close to DMSO values. Two classifiers were then built, one for each dataset.
  • a subset of 12 out of the 20 genes in the signature (Example 5, Table 3; version 1.0 of the classifier) was chosen in an iterative process similar to the one described for version 1.0.
  • each subset was used for classifier training and the classifiers' performance was recorded (using the 'leave compound out' method). The number of times each gene participated in classifiers with top 20% performance was counted, and those genes with significantly higher than average counts were selected to be in the final gene list.
  • Random Forest classifier was trained for each dataset using only this gene subset.
  • the Random Forest parameters (number of trees, mtry, class weight, score threshold) were optimized as described for classifier- version 1.0 (Example 5).
  • the prediction for a compound was decided by a majority of sample predictions for the compound. Compounds with the same number of toxic and non-toxic predictions were counted as half a successful prediction.
  • the classifier performance of the resulting classifiers was 90.9% (30/33) sensitivity and 78.6% (11/14) specificity for the classifier with no DMSO transformation, and 86.4% (28.5/33) sensitivity and 89.3% (12.5/14) specificity for the classifier with DMSO transformation.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Immunology (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Hematology (AREA)
  • Urology & Nephrology (AREA)
  • Organic Chemistry (AREA)
  • Analytical Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Microbiology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Biotechnology (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Zoology (AREA)
  • Toxicology (AREA)
  • Wood Science & Technology (AREA)
  • Cell Biology (AREA)
  • Medicinal Chemistry (AREA)
  • Pathology (AREA)
  • General Physics & Mathematics (AREA)
  • Food Science & Technology (AREA)
  • Tropical Medicine & Parasitology (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • Genetics & Genomics (AREA)
  • Gastroenterology & Hepatology (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)

Abstract

L'invention concerne des méthodes de prévision de l'hépatotoxicité d'un composé par la fourniture d'un système de culture d'hépatocytes in vitro capable d'exprimer un panel de gènes, la culture pouvant être incubée avec le composé ou les composés à analyser, et le niveau d'expression du panel de gènes pouvant être mesuré. En comparant le niveau d'expression du panel de gènes dans le système de culture exposé au(x) composé(s) analysé(s) et l'expression génique du même panel de gènes dans un système de culture de contrôle, l'hépatotoxicité du composé ou des composés peut être prédite.
PCT/IL2016/050566 2015-06-03 2016-06-02 Méthodes de prévision de l'hépatotoxicité Ceased WO2016193977A2 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IL239188A IL239188A0 (en) 2015-06-03 2015-06-03 Methods for predicting hepatotoxicity
IL239188 2015-06-03

Publications (2)

Publication Number Publication Date
WO2016193977A2 true WO2016193977A2 (fr) 2016-12-08
WO2016193977A3 WO2016193977A3 (fr) 2017-03-16

Family

ID=55022871

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IL2016/050566 Ceased WO2016193977A2 (fr) 2015-06-03 2016-06-02 Méthodes de prévision de l'hépatotoxicité

Country Status (2)

Country Link
IL (1) IL239188A0 (fr)
WO (1) WO2016193977A2 (fr)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110003320A (zh) * 2019-03-22 2019-07-12 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体bp1-7及应用
CN110054676A (zh) * 2019-03-07 2019-07-26 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体ap3-5及应用
CN110054674A (zh) * 2019-03-07 2019-07-26 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体ap2-19及应用
CN110054688A (zh) * 2019-03-07 2019-07-26 深圳市龙华区人民医院 一种抗ttc36单克隆抗体及其应用
CN110517790A (zh) * 2019-06-24 2019-11-29 江苏大学 基于深度学习和基因表达数据的化合物肝毒性早期预测方法
WO2019237015A1 (fr) * 2018-06-08 2019-12-12 The Trustees Of Columbia University In The City Of New York Procédés et systèmes de prédiction de taux de réussite d'essais cliniques

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040005547A1 (en) * 2002-03-14 2004-01-08 Franziska Boess Biomarkers and expression profiles for toxicology

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019237015A1 (fr) * 2018-06-08 2019-12-12 The Trustees Of Columbia University In The City Of New York Procédés et systèmes de prédiction de taux de réussite d'essais cliniques
CN110054676A (zh) * 2019-03-07 2019-07-26 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体ap3-5及应用
CN110054674A (zh) * 2019-03-07 2019-07-26 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体ap2-19及应用
CN110054688A (zh) * 2019-03-07 2019-07-26 深圳市龙华区人民医院 一种抗ttc36单克隆抗体及其应用
CN110054688B (zh) * 2019-03-07 2020-10-27 深圳市龙华区人民医院 一种抗ttc36单克隆抗体及其应用
CN110054676B (zh) * 2019-03-07 2022-08-02 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体ap3-5及应用
CN110054674B (zh) * 2019-03-07 2022-08-05 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体ap2-19及应用
CN110003320A (zh) * 2019-03-22 2019-07-12 深圳市龙华区人民医院 免疫原性多肽以及抗ttc36抗体bp1-7及应用
CN110517790A (zh) * 2019-06-24 2019-11-29 江苏大学 基于深度学习和基因表达数据的化合物肝毒性早期预测方法

Also Published As

Publication number Publication date
IL239188A0 (en) 2015-11-30
WO2016193977A3 (fr) 2017-03-16

Similar Documents

Publication Publication Date Title
Bunney et al. Microarray technology: a review of new strategies to discover candidate vulnerability genes in psychiatric disorders
Gianmoena et al. Epigenomic and transcriptional profiling identifies impaired glyoxylate detoxification in NAFLD as a risk factor for hyperoxaluria
Chen et al. Sertraline induces endoplasmic reticulum stress in hepatic cells
WO2016193977A2 (fr) Méthodes de prévision de l'hépatotoxicité
Greening et al. Chronic methamphetamine interacts with BDNF Val66Met to remodel psychosis pathways in the mesocorticolimbic proteome
Datson et al. Expression profiling in laser‐microdissected hippocampal subregions in rat brain reveals large subregion‐specific differences in expression
US20090208939A1 (en) Identification of Molecular Diagnostic Markers for Endometriosis in Blood Lymphocytes
Li et al. Time-dependent changes in transcriptional profiles within five rat brain regions in response to nicotine treatment
Chorley et al. Methodological considerations for measuring biofluid-based microRNA biomarkers
Antoine et al. Safety biomarkers for drug-induced liver injury–current status and future perspectives
US20150225800A1 (en) Methods and assays for treatment of bladder cancer
Raza et al. qPCR Analysis reveals association of differential expression of SRR, NFKB1, and PDE4B genes with type 2 diabetes mellitus
JP2008512984A5 (fr)
Blaveri et al. Expression profiling of a genetic animal model of depression reveals novel molecular pathways underlying depressive-like behaviours
Mitra et al. Analysis of the toxicogenomic effects of exposure to persistent organic pollutants (POPs) in Slovakian girls: correlations between gene expression and disease risk
KR101604178B1 (ko) 약물에 의한 간 손상의 예측 및 진단을 위한 바이오마커
Dimopoulou et al. Flusilazole induces spatio-temporal expression patterns of retinoic acid-, differentiation-and sterol biosynthesis-related genes in the rat Whole Embryo Culture
Galvin et al. Expression profiling in the aging brain: a perspective
Dimopoulou et al. A transcriptomic approach for evaluating the relative potency and mechanism of action of azoles in the rat Whole Embryo Culture
CN102959090B (zh) 用于表征和鉴别遗传毒性化合物的基因表达分析
Kleinjans Toxicogenomics-based cellular models: Alternatives to animal testing for safety assessment
WO2017082943A1 (fr) Articles servant à diagnostiquer une fibrose hépatique
US20220136065A1 (en) High-grade serous ovarian carcinoma (hgsoc)
JP2021099351A (ja) 生体サンプル中のミトコンドリア複製の機能不全をインビトロで調査するための方法、そのキット及び使用、早老症様の症候群又は症状に対する治療方法、並びに特定のプロテアーゼ阻害剤及び／又はニトロソ酸化還元ストレススカベンジャー化合物を同定するためのスクリーニング方法
Lee-Ødegård et al. Serum proteomic profiling of physical activity reveals CD300LG as a novel exerkine with a potential causal link to glucose homeostasis

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16774554

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 09.04.2018)

122 Ep: pct application non-entry in european phase

Ref document number: 16774554

Country of ref document: EP

Kind code of ref document: A2