Pulsed Of Silac Proteomic Quantitative Methods Of Optimization-based Microrna Target Gene Screening Applications

Tremendous progresses have been made in quantitative proteomics during the past few years. Among them stable isotope labeling with amino acids in cell culture (SILAC) is widely used in biomedical research because of its high quantitative accuracy and high resolution. The goal of this study is to optimize and apply a quantitative proteomic strategy to screen potential targets of miR-128 in glioma cells based on pulsed SILAC. In addition, we compared the effectiveness of various data acquisition methods for hybrid linear ion trap fourier-transform ion cyclotron resonance mass spectrometer (LTQ-FT) when analyzing peptide mixtures with different complexities.Glioma is the most aggressive and malignant primary brain tumor. Even with the most up-to-date surgery followed by concomitant radiation and chemotherapy, the median survival is short (always less than14 months). There is a recognized need for new approaches based on increased understanding of the biological and molecular nature of this tumor. Recent studies have demonstrated an important role for miRNAs in tumorigenesis, which function as tumor suppressors or oncogenes by regulating the expression of their target genes. Interestingly, miR-128 was found to be down-regulated in gliomas. Bioinformatics prediction and experimental results indicate that miR-128 can target E2F3a or BMI-1and inhibit cell proliferation in gliomas. However, the other targets of miR-128 remain unknown. The majority of reports describing identification of miRNA targets are based on computational prediction or experimental methods. The computational prediction always raises false positive results and experimental approach is limited by low-throughput. It is reported that miR-128 have about 2000 potent targets in glioma cells. Therefore, it is essential to develop the high-throughput experimental approaches to verify these targets. Pulsed SILAC is a high-throughout SILAC method to screen the potential targets after microRNA transfection. In this approach, labeling is restricted to newly synthesized proteins, permitting determination of the impact on protein synthesis rates. But this approach may get false negative results because of the high background protein abundance when labeling newly synthesized proteins with medium isotope-labelled amino acids. We then modified and optimized this quantitative proteomic strategy and systematically identify potent targets of miR-128 in glioma cells for better understanding of its pathogenesis.This dissertation consists of three parts and the contents are summarized as follows: the status quo of quantitative proteomics, including SILAC and pSILAC, are reviewed in the first chapter. The aggressive and malignant features of glioma and the needs for new approaches to understand glioma pathogenesis are also introduced.The work in chapter two described the establishment and optimization of a quantitative proteomic strategy based on pulsed SILAC and LTQ-FT to screen potential targets of miR-128 in glioma cells. Totally 1897 proteins were identified and 1459 were quantified in duplicate experiments. 133 proteins were down-regulated by at least 30% after miR-128 transfection. Among these down-regulated proteins in miR-128 transfected glioma cells, CCNK,TPP2, MCCC2,UBA2,NNMT,ARL8B,SFRS1,PFKL,GLTP and PHGDH were verified down-regulated in luciferase reporter experiments, indicating that they may be the potential targets of miR-128. We also discussed the criterion of choosing suitable potent targets for validation with biological approaches and found that better results could be obtained by manually checking the raw mass spectrum respectively. In addition, we constructed the interaction network for proteins down-regulated by at least 15% in miR-128 transfected glioma cells with STRING 8.2 and Cytoscape to predict more potential targets, and found that ribosomal proteins and translation initiation factors (EIF) were the dominant nodes. This result indicated that these ribosomal proteins and translation initiation factors may be potential targets too.The work in chapter three is to compare the effectiveness of various data acquisition methods in shotgun proteomics. We used a LTQ-FT mass spectrometer to analyze two peptide mixtures with different complexity by different acquisition strategies. For the peptide mixture of four standard proteins, sequence coverage obtained by selected ion monitoring scans (SIM3) was 1.51 to 1.9 fold of that by survey MS scans (FT10), respectively. For the peptide mixture of yeast proteins, the strategy of only acquiring double and triple charged ions (FT₂3) got 64.1% more MS/MS spectra identifications than that of total acquisition of single, double and triple charge ions (FT₁23). Finally, the features of the spectra obtained with different acquisition mode were also compared. In conclusion, optimal acquisition methods are needed for samples with different complexity.