UNDERSTANDING PHYSIOLOGY OF DISEASES AND CELL LINES USING OMICS BASED APPROACHES

OMICS;Bioinformatics;Proteomics;Chemical & Biomolecular Engineering

Kumar, AmitYarema, Kevin J ;

Johns Hopkins University

关键词: OMICS;    Bioinformatics;    Proteomics;    Chemical & Biomolecular Engineering;   

Others  :  https://jscholarship.library.jhu.edu/bitstream/handle/1774.2/60269/KUMAR-DISSERTATION-2015.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: JOHNS HOPKINS DSpace Repository

【 摘 要 】

This thesis focuses on understanding physiology of diseases and cell lines using OMICS based approaches such as microarrays based gene expression analysis and mass spectrometry based proteins analysis. It includes extensive work on functionally characterizing mass spectrometry based proteomics data for identifying secreted proteins using bioinformatics tools. This dissertation also includes work on using omics based techniques coupled with bioinformatics tools to elucidate pathophysiology of diseases such as Type 2 Diabetes (T2D).Although the well-known characteristic of T2D is hyperglycemia, there are multiple other metabolic abnormalities that occur in T2D, including insulin resistance and dyslipidemia. In order to attain a greater understanding of the alterations in metabolic tissues associated with T2D, microarray analysis of gene expression in metabolic tissues from a mouse model of pre-diabetes and T2D to understand the metabolic abnormalities that may contribute to T2D was performed. This study also uncovered the novel genes and pathways regulated by the insulin sensitizing agent (CL-316,243) to identify key pathways and target genes in metabolic tissues that can reverse the diabetic phenotype. Specifically, he found significant decreases in the expression of mitochondrial and peroxisomal fatty acid oxidation genes in the skeletal muscle and adipose tissue of adult MKR mice, and in the liver of pre-diabetic MKR mice, compared to healthy mice. In addition, this study also explained the lower free fatty acid levels in MKR mice after treatment with CL-316,243 and provided biomarker genes such as ACAA1 and HSD17b4.Using results from T2D microarrays studies, a multi-tissue computational model was created using metabolic reconstructions for in silico simulation of T2D for a better understanding of the disease pathophysiology. A time-efficient algorithm for generating tissue-specific metabolic models was presented in this study. The flux balance analysis using the multi-tissue model showed that the degradation pathways of branched-chain amino acid and fatty acid oxidation were significantly downregulated in MKR mice versus healthy mice. The T2D multi-tissue model was able to explain the high level of branched-chain amino acids and free fatty acids in plasma of T2D subjects from a systems level metabolic fluxes perspective.In addition to T2D studies, this dissertation also reports identification of the complete collection of proteins which make up the Chinese hamster ovary (CHO) cells proteome which has been an invaluable source of information for scientists, allowing them to engineer their cell lines to increase the efficiency of therapeutics production. Proteomics has been especially attractive for biotechnology applications since it can provide an understanding of disease states and aid drug discovery and development. Moreover, CHO cells are the preferred host cell line for manufacturing a variety of biologicals including monoclonal antibodies. A proteomics and bioinformatics analysis on the spent medium from CHO cells was performed. From the analysis of supernatant of post-centrifugation CHO cells, identification of thousands of unique proteins that are potentially secreted from the CHO cells was done. In order to categorize these proteins functionally, multiple bioinformatics tools including SignalP, TargetP, SecretomeP, TMHMM, WoLF PSORT, and Phobius were implemented. This analysis provided information on the cellular localization of the proteins found in the supernatant, including the presence of transmembrane domains and signal peptides. Proteins were shown to be localized to the secretory pathway, including ones playing role in cell growth, proliferation, and folding as well as those involved in degradation and removal of other proteins. As a part of this effort, a publically accessible web-based tool called GO-CHO (http://ebdrup.biosustain.dtu.dk/gocho/) was created to functionally categorize the proteins. This work and database will enable the CHO community to rapidly identify high abundance host cell proteins in their cultures in order to facilitate processing and purification efforts in the future. Moreover, the compartmentalization strategies presented in this work will help the CHO community in understanding CHO secretory machinery.