The Study Of Postgenomics: Proteomic Analysis Of Morphine Dependence Development Of Technology For DNA Methylome Analysis

With the completion of Human Genome Project (HGP) the genomic study has moved from genomic sequencing to functional study on genome-scale, emphasizing exploration of the expression and regulatory network of genes on genome-wide. The diversified post-genomic field includes, but not limits, genomic SNP analysis, transcriptomics, proteomics, and newly developed epigenomics.Proteomics is to study protein expression and interaction of entire proteins in specific tissues and cells, and recently has been intensively used in human neurological diseases such as Parkinson's disease, Alzheimer disease, drug addiction and schizophrenia. Drug abuse is recognized as long-term adaptation in brain which is resulted from repeatedly usage of psychoactive substances. Therefore, drug addiction is also presumed as chronic and relapsing encephalopathy. The numerous lines of evidence have suggested that dopaminergic and glutamatergic neuronal synaptic plasticity play an important role in drug addition. In general, psychoactive substances strongly affect on the neuronal systems through dopamine D-1 and glutamate N-methyl-D-aspartate (NMDA) andα-amino-3-hydroxy-5-methylisoxazole -4-propionic acid (AMPA) receptors, which induce intracellular transcriptional and translational cascades, leading to adaptive changes in gene expression, protein expression and molecular constitute. The long-term adaptation of synaptic plasticity induced from the above molecular alteration, causes the alteration of synaptic morphology and structure which further result in irreversible neural dysfunction.Prefrontal cortex (PFC) is one of important brain regions in reward circuits, which is critical for many cognitive functions involving inhibitory control and decision making. As the relationship between a cell's chemistry and structure becomes increasingly apparent, it is not surprised that chronic exposure to drugs of abuse has been shown to alter the morphology of neurons in reward circuits of the brain. Chronic morphine administration decreases the size and caliber of dopamine neurons in ventral tegmental area (VTA). In addition, chronic morphine administration has been shown to decrease the complexity of dendritic branching and the number of spines on medium spiny neurons in the nucleus accumbens (NAc) and PFC. For better understanding the synaptic molecular mechanism involved in addiction, we performed proteomic analysis of PSD fraction of rats' PFC in three phases of morphine induced conditioned place preference (CPP). The first part of the thesis is to present our proteomic analysis of PSD fraction of rats' PFC in three phases of morphine induced CPP. The effect of morphine dependence is evaluated by CPP model, the procedures of which include nature preference test, conditioned acquisition, conditioned extinction and conditioned reinstatement. Next we analyzed the expression profiling of PSD fraction of rats' PFC in three phases of CPP with two-dimensional gel electrophoresis (2-DE). Furthermore, we identified differential expression proteins with MALDI-TOF-MS and validated the results of proteomics with western blot.Results:1. Morphine induced CPP model was established, including acquisition, extinction and reinstatement phases. The rats showed strong preference to morphine-paired chamber in acquisition test after morphine conditioning. The morphine-induced preference is diminished after saline-paired extinction training, and could be re-established after morphine priming injection.2. Proteomic analysis of PSD fraction of rats' PFC in three phases of morphine induced CPP with 2-DE resulted in 80 differential proteins, of which protein cyclic nucleotide phosphodiesterase 1 (CNPasel) showed significantly differential expression in the three phases of CPP respectively, and the remaining proteins were altered in one or two phases of CPP.3. The differential proteins were identified with MALDI-TOF-MS spectrum, of which 21 proteins were successfully identified in acquisition phase, 14 in extinction phase and 23 in reinstatement phase. Based on bioinformatics analysis, 58 identified differential proteins are found to involved in many vital biological functions, such as energy metabolism, signal transduction, synaptic transmission, cytoskeleton proteins, molecular chaperones, and synaptic proteins synthetic machine.4. The expression of representative protein CNPasel was verified for assessment of the accuracy of proteomic results with western blot. The result showed that there were 2.1 fold elevation in the acquisition phase, 1.6 fold decrease in the extinction phase and 1.4 fold increase in the reinstatement phase, which were consistent with the analysis of 2-DE and silver staining. The data indicates that the changes in protein expression identified by proteomic analysis of synaptic fraction of PFC in the three phases of CPP are reliable.Conclusion:The proteomic analysis of PSD fraction of rats' PFC in the three phases of morphine induced CPP resulted in identification of many differentially expressed proteins. Based on bioinformatics analysis, these proteins are classified into many important biological functions, such as energy metabolism, signal transduction, synaptic transmission, cytoskeleton proteins, molecular chaperones and synaptic proteins synthetic machine. Therefore, our study, in the first time, revealed the expression profiling of PSD fraction of rats' PFC under morphine induced conditioned dependence including acquisition, extinction and reinstatement, which provides many crucial candidates for further study of molecular synaptic function in morphine addiction.Epigenetics is to study heritable changes, such as DNA methylation and histone modification that modulate chromatin organization and gene expression in the absence of change in DNA sequence. The most widely studied epigenetic modification in humans is the cytosine methylation of DNA within the dinucleotide CpG. DNA methylation has been shown to involve in many biological processes, such as gene silencing, transposable inhibition, X-chromosome inactivation and cell differentiation. An altered pattern of DNA methylation has been associated with etiology of some severe human diseases such as cancer, mental health and immunological diseases.A range of approaches are available for assessing DNA methylation, such as sodium bisulfite conversion of genomic DNA combined with sequencing, methylation specific digestion, methylation specific PCR (MSP) and DHPLC. All of the above approaches are limited to study DNA methylation statue on gene-by-gene basis. Obviously, these small-scale approaches can not meet the task to identify epigenetic modification at the genomic level in the complex human diseases. This results in the development of new field, namely epigenomics. Epigenomics is defined as studying epigenetic modifications on genome-scale. DNA methylome analysis is one of important direction of Epigenomics study. Although a few progresses in epigenomics have been achieved, there is still much difficulty to be overcome for developing a reliable and accurate methodology to identify DNA methylome. Therefore, in this project we developed a high through-put 9K CpG island (CGI) array which could be potentially applied into study of underlying DNA methylation modification in a variety of biological processes and common human diseases.Although 9K CGI array provides an efficient tool for DNA methylation profiling analysis, the bottleneck of DNA methylome study is genome-wide methylated DNA enrichment. Thus, maximal enrichment of methylated DNA is essential in subsequent detection of differential methylation CGI based on high through-put technologies. Most commonly used genome-wide methylated DNA enrichment approaches including methods based on methylation sensitive restriction enzyme digestion -differential methylation hybridization (DMH) and microarray based methylation assessment of single samples (MMASS), as well as ones based on protein affinity purification - methylcytosine binding protein (MBD protein) affinity purification and methylated DNA immunoprecipitation (MeDIP). Previous study has demonstrated that MeDIP is more sensitive than MBD purification for detecting methylated DNA, nevertheless, a systematic comparison of methylation sensitive restriction enzymes digestion based assays including DMH and MMASS, with MeDIP haven't been performed. Therefore, we first optimized several important experimental parameters of these three methods, and then performed evaluation of sensitivity and accuracy among these three methods. Finally, we performed GO analysis of differential CGI related genes from methylation profiling of Ges-1 and MGC-803 cell lines with MMASS and MeDIP assays, and the results showed lots of genes involved in important biological functions and processes.The second part of the thesis describes our effort in developing human CGI array for DNA methylome analysis as well as optimizing and evaluating methylated DNA enrichment approaches based on the CGI array. We identified CGI clones through sequencing of human CGI library by ourselves and Wellcome Trust Sanger Institute, respectively. In addition, we predicted CGIs from the region of the first exon and 10kb upstream of TSS of some cancer related genes and imprinted genes. All the representative CGI clones and predicted CGIs of target genes were amplified with PCR for CGI array spotting. In addition, we optimized several important parameters of DMH, MMASS and MeDIP, including PCR annealing temperatures and antibody incubation time in immunoprecipitation through methylation profiling of gastric adenocarcinoma cell line MGC-803 with 9K CGI array. Furthermore, we performed a systematic evaluation of sensitivity and accuracy of the three assays through methylation profiling of gastric epithelium cell line Ges-1 and adenocarcinoma cell line MGC-803 with 9K CGI array. Finally, we carried out GO analysis of differential CGI related genes from methylation profiling of Ges-1 and MGC-803 cell lines with MMASS and MeDIP assays. Results:1. 18,816 clones (18k) from human CGI library were sequenced. In addition, 12,192 clones identified from the same library bought from Wellcome Trust Sanger Institute for developing high through-put CGI array. After sequences aligning, blasting and clustering, 10,749 representative clones were selected from 18K dataset with the redundancy of 37.02%, as well as 6,445 representative clones from 12K dataset with the redundancy of 37.84%.2. Based on bioinformatics prediction, 33,116 CGI were identified from whole human genome with CpGi130. 10,749 representative clones selected from 18K dataset and 6,445 representative clones from 12K dataset were matched with predicted CGIs from whole human genome. The overlapped clones were defined as CGI clones, otherwise as non-CGI clones. Totally, 8,967 clones were matched to our bioinformatics prediction and identified as representative CGI clones. In addition, 830 unique CGIs were predicted from the region of the first exon and 10kb upstream of 450 cancer related genes and 90 imprinted genes with CpGi130. We annotated all the representative CGI clones from CGI library and CGIs predicted from target genes including the location on the chromosome, the adjacent genes, the position relationship with genes and the description of genes.3. 8,967 representative CGI clones from CGI library and 830 CGIs from target genes were amplified, of which 9,223 CGI probes were successfully amplified for 9K CGI array spotting. The evaluation of 9K CGI array manifests that both of the signal and the consistency of hybridization are very high.4. Several important parameters were optimized in three methylated DNA enrichment assays including DMH, MMASS and MeDIP. The comparison of different PCR annealing temperatures showed that relative high temperature (72℃) for amplification of products, resulted from methylation-sensitive and methylation-dependent digestion, significantly improved the sensitivity of assessment of differential methylation of CGIs. In addition, the comparison of different antibody incubation time showed that the relative long time of the primary antibody incubation (12h) and the suitable time of the secondary antibody incubation (3h around) increased the efficiency of methylated DNA enrichment and, in the meantime, decreased the unspecific binding of unmethylated DNA.5. A systematic evaluation of the sensitivity and accuracy of the three enrichment methods among DMH, MMASS and MeDIP were performed. The results show that: 1) the three methods identify some common differential methylated CGI probes, however, each method detected unique differential targets that were ranked very low in the other two; 2) MMASS is the highest in sensitivity, MMASS is similar with MeDIP in accuracy and DMH is the lowest either in sensitivity or in accuracy among the three assays.6. GO analysis of 361 differential CGI related genes from methylation profiling of Ges-1 and MGC-803 cell lines with MMASS and MeDIP assays was performed. The results showed that the genes were involved in many vital biological processes including cellular component, molecular function and biological process. Conclusion:We developed human 9K CGI array with 9223 CGI probes identified from human CGI library and target genes, which could provide a high through-put tool for DNA methylome analysis on genome-wide. The optimization of several important parameters in DMH, MMASS and MeDIP significantly improves the sensitivity and specificity of the assessment of DNA methylation profiling. Furthermore, the systematic evaluation suggests that MMASS is the highest in sensitivity among these three assays, nevertheless both enzyme digestion based method - MMASS and immunoprecipitation based method - MeDIP are complementary in detection of differential methylated candidates on genome-scale. Therefore, employment combining two or more enrichment approaches could result in better and complete assessment of DNA methylome. In addition, the results of GO analysis suggest the differential CGI related genes which are potentially regulated by differential methylation are involved in lots of vital biological processes and functions. The screening based on 9K CGI array could provide many important candidates for the study of epigenetic mechanism in cancer, and 9K CGI array will be very promising in DNA methylation profiling of human physiological and pathological samples.