Clustering Analysis Of Gene Express Profiles Of Mouse Primary Culture Hepatocyte Induced By 13 Chemicals

The great development of DNA microarray techniqe and Genomics had promoted the birth of Toxicogenomics. It is a new sub-discipline into which new technologies of bioinfomatics and Genomics had been incorporated. Toxicogenomics is the study of the relationship between the structure and activity of the genome (the cellular complement of genes) and the adverse biological effects of exogenous agents. Firstly, it will help us understand the relationship between the polymorphism of gene and the toxicity of chemicals. Secondly, it will help us understand the relationship between toxicity of chemicals and gene express on the level of gene express, so that we can understand the mechansim of toxic effect better. Finally, Toxicogenomics make it possible that classification of chemicals based on gene express profiles. We conceived to investigate the gene express profiles of mouse primary culture hepatocyte induced by 13 chemicals by using DNA microarray, by the aid of clustering analysis on the gene express profiles, find out the special "gene fingerprints" of chemicals and establish a database, and then try to classify chemicals based on the level of response of genome, so that make the classification of chemicals been relative to its gene express profiles. At the same time, attempted to use the toxicological microarray which we designed on the classification of chemicals. The main results were summed up as follows:1. The manufacture of toxicological microarrayWe cooperated with HongKong University to develop mouse toxicological microarray.1.1 1796 gene were choosed from NLA 15K and 7.4K mouse cDNA library to form a toxicological microarray. Those are concerned to eight functional groups: cell growth and maintenance, extracellular matrix, signal transduction, stress, metabolism, transcription, translation and transport.1.2 The PCR amplification were done on 1796 gene. The qualification rate of productsof PCR amplification were 97.57%. This shows that the PCR amplification of choosed gene were successful.1.3 PicoGreen fluorescence staining were done on the microarray to investigate sample application. The fluorescence on the microarray were even and strong. It proved that the quality of microarray were fine and can be used in next experiment.1.4 The coefficient of variation of positive control were all lower than 0.15. This indicated that the change range of microarray results were small and the results were reliable.1.5 The Pearson correlation coefficient of two microarray hybridization test in same batch and different batch were 0.907 and 0.887 respectively. The coincidence rate of two microarray hybridization test in same batch and different batch were 100%. These indicated that two microarray hybridization test in same batch and different batch had good repeatability.1.6 False positive rate of self-tset were 0.56%(cutoff: 2.0 and 0.5), were closed to lower limit as reported(0.5%~3%). This proved that the microarray system were steady and reliable.2. The establishment of cell model2.1 Mouse hepatocytes were isolated using a two-step perfusion method as described (Klaunig, et al. 1981). Normal yields per liver were than 107 viable cells. The viability is usually 85-95%. Hepatocytes grows fine after inoculation.2.2 Detected cytotoxicity of 13 chemicals by using MTT assay. Dose response curve were drawn. LC30 of chemicals were calculated respectively, which been used in the study of toxicological gene express profiles as usage dose.3. Analysis Of Gene Express Profiles Induced By 13 Chemicals3.1 Total RNA of sample were extracted by RNA Isolation Solution which been confected by ourselves. The RNA were pure enough to be applied directly for microarray hybridization.3.2 There were total 580 differentially expressed gene induced by 13 chemicals on three time point, in which the amount of differentially expressed gene on all three time point were 289, the amount of differentially expressed gene on 4H and 24H were 316, on 4H and 72H were 313, on 24H and 72H were 386. As a whole, the amount of differentiallyexpressed gene increased gradually by time.3.3 Validation of microarray experiment were done on 4 gene in three pairs samples which were selected randomly by using Real-Time Fluorescence Quantitative PCR. The results of validation were according with microarray experiment. It indicated that the results of microarray were reliable.3.4 Clustering analysis of gene express profiles by using average-linkage Hierarchical clustering. The results as followed:a) The 4H's classification of chemicals based on gene express profiles were great different with classification based on chemical structure. We presumed the reason may be that 4H' s gene express profiles could not represent the speciality of chemicals toxicity.b) The 24H's classification of chemicals based on gene express profiles were closed to 72H's. But the 4H's classification were different with 24H's and 72H's evidently. It indicated that the gene express profiles may be steady among limited time.c) DMBA and DBA were clustered into one cluster on all three time point, and their relationship were most tightest among all chemicals. It suggested that their toxical structural base may be the four membered ring.d) The 24H's and 72H's classification of chemicals based on gene express profiles were near to the classification based on chemical structure, the coincidence rate were 76.9% and 61.5% respectively.Our study indicated that: classifying chemicals by gene express profiles, so that to make classification relative to chemical toxic action, and then to study their toxic action mechanism and to predict the toxicity of unknown chemicals is a worthy and hopeful study direction; Our study also provided experimental basis for the establishment of new method to detect enviromental expose.