http://hdl.handle.net/1721.1/7794 Tue, 21 May 2013 11:22:35 GMT 2013-05-21T11:22:35Z http://hdl.handle.net/1721.1/78438 Rational design and directed evolution of probe ligases for site-specific protein labeling and live-cell imaging White, Katharine Alice Chemical fluorophores have superior photophysical properties to fluorescent proteins and are much smaller. However, in order to use these probes for live-cell protein imaging, highly specific labeling methods are required. Here, we will describe three efforts to re-engineer the E. coli enzyme, lipoic acid ligase (LplA), to catalyze the ligation of small-molecule probes onto recombinant proteins. We call this collection of methods the PRIME (PRobe Incorporation Mediated by Enzymes) methodologies. First, we describe the structure-guided mutagenesis of LplA and the identification of an LplA variant that can ligate a blue coumarin fluorophore onto a 13-amino acid LplA acceptor peptide (LAP2). This "coumarin ligase" can be used to image cellular proteins with high specificity, sensitivity, and minimal perturbation of the biology of the protein of interest. We also demonstrate how subpopulations of a protein of interest can be labeled using genetically targeted coumarin ligase. Second, we describe our attempts to use yeast display evolution and fluorescence activated cell sorting (FACS) to evolve a truncated LplA enzyme. The original truncated enzyme had severely decreased activity for LplA's natural substrate, lipoic acid. We created a 107 library of LplA mutants and, after four rounds of selection, produced a truncated LplA mutant with lipoylation activity equivalent to full-length LplA. We next sought to evolve activity for an unnatural small molecule probe, but found that this strategy was limited by both increased hydrophobic probe sticking when using the truncated enzyme and some enzyme-dependent nonspecificity. Finally, from a library of 107 LplA mutants, we evolved a full-length LplA capable of ligating an unnatural picolyl azide (pAz) substrate. We demonstrated improved activity of the "pAz ligase" in the secretory pathway and cell surface, two regions where coumarin ligase is inactive. This enzyme can also be used to image cell surface protein-protein interactions as well as label proteins as they are trafficked through the endoplasmic reticulum. These probe ligases will be useful tools for cell biologists interested in studying protein function or protein-protein interactions in the context of living cells. Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemistry, 2012.; Cataloged from PDF version of thesis.; Includes bibliographical references. Sun, 01 Jan 2012 05:00:00 GMT http://hdl.handle.net/1721.1/78438 2012-01-01T05:00:00Z http://hdl.handle.net/1721.1/78437 Understanding the functions of HMGB proteins in the mechanism of action of cisplatin Park, Semi, Ph. D. Massachusetts Institute of Technology High mobility group box (HMGB) proteins are DNA-binding proteins that regulate many important DNA-related processes. They are known to recognize the major lesion present in cisplatin-modified DNA, and have been assumed to increase cisplatin cytotoxicity by "shielding" the damaged site from the cellular repair apparatus. This thesis will describe the work in the area of molecular biology and biochemistry to improve our understanding of the functions of HMGB proteins in the mechanism of action of cisplatin. In Chapter 1, the molecular biology of cisplatin and HMGB proteins is described based on previously reported in vivo and in vitro experiments. The interaction of HMGB proteins and platinated DNA is reviewed based on several structural studies of platinum-DNA adducts of cisplatin, HMG box motif, and the binding complex of the two. Several cell-based studies supporting a repair-shielding hypothesis, suggesting an inhibitory function of HMGB proteins in the repair of cisplatin damage will be introduced with a focus on HMGB1, the most vigorously investigated HMGB protein. Additionally, other HMGB 1-mediated processes that may be related to cisplatin-triggered cell death will be discussed. Lastly, the correlation between testis-specific HMGB proteins and the cisplatin hypersensitivity of testicular germ cell tumors will be discussed. Although some HMGB proteins including HMGB 1 inhibit the repair of platinated damage on DNA in vitro, experiments conducted in live cells reveal conflicting correlation between the expression level of HMGB1 and cisplatin cytotoxicity. Chapter 2 describes studies in cultured human cancer cells aimed at examining the intracellular repair of platinum-modified DNA and the influence of HMGB1 on the repair processes. The expression of HMGB1 is artificially down-regulated by RNAi. HeLa and A549 cell lines present different trends in cisplatin cytotoxicity upon HMGB1 knockdown. Intracellular repair of cisplatin is lower in knockdown cells regardless of the parental cell line. This result stands in opposition to what is expected from the repair-shielding hypothesis. In addition, the repair of different cisplatin adducts was investigated in fibroblast cells deficient in one of the nucleotide excision repair proteins in order to understand the repair mechanisms of each adduct. Chapter 3 presents an in vitro study investigating the redox-dependence of the binding interaction of HMGB1 to the cisplatin-1,2-d(GpG) intrastrand cross-link. Two cysteine residues in HMGB1 domain A form a reversible disulfide bond under mildly oxidizing conditions. Both HMGB 1 domain A and full-length HMGB 1 presented significantly weaker binding to a DNA probe containing a 1,2-d(GpG) intrastrand cross-link when in their oxidized state compared to their fully-reduced form. Mutagenesis studies on the cysteine residues revealed that this redoxdependence originates from disulfide bond formation. Footprinting analysis of a platinated DNA probe bound to oxidized or reduced domain A showed that the asymmetric binding mode of domain A to platinated DNA is not significantly altered by oxidation. These results suggest that the cellular redox environment can influence the interaction of HMGB 1 with the platinated DNA. In Chapter 4, the binding properties and repair inhibitory function of HMGB4 to cisplatin-modified DNA are described. Based on its testis-restricted expression profile and sequence similarity with HMGB 1, we propose that HMGB4 functions as a cisplatin cytotoxicity enhancer in TGCT. To verify this hypothesis, HMGB4 and its binding domains were generated recombinantly and interactions with cisplatin-modified DNA were investigated in vitro. The binding properties of HMGB4 are quite similar to those of HMGB 1 except for a few differences: i) full-length HMGB4 has stronger binding affinity than full-length HMGB1, because of a lack of a C-terminal acidic tail. ii) binding of HMGB4 domain A is much more symmetric with respect to the platinated lesion than that of HMGB 1 domain A. Furthermore, HMGB4 presented stronger repair inhibition capacity than HMGB1 at an equimolar concentration. These results support the hypothesis that HMGB4 enhances cisplatin cytotoxicity in TGCT. Chapter 5 will be the conclusion chapter of this thesis. Works in this thesis will be summarized with what we can learn from those results. In additions, future directions in the study of HMGB proteins will be suggested. Appendices A and B describe the incomplete work on HMGB4 in live cells. Appendix A delineates the expression profile of human HMGB4 established by western blot analysis. Human HMGB4 presented a testis-preferred expression. In Appendix B, attempts to establish a HMGB4 knockdown testicular cancer cell line are described. Thesis (Ph. D. in Inorganic chemistry)--Massachusetts Institute of Technology, Dept. of Chemistry, 2012.; Cataloged from PDF version of thesis. Vita.; Includes bibliographical references. Sun, 01 Jan 2012 05:00:00 GMT http://hdl.handle.net/1721.1/78437 2012-01-01T05:00:00Z http://hdl.handle.net/1721.1/78361 Computational analyses of immune responses at disparate temporal and spatial scales Wolfson, Mikhail Yanislavovich In order to perform reliably and protect against unpredictable attackers, immune systems are organized via complex, hierarchical cooperativity. This organization is necessary for their function and a tremendous challenge to their understanding that has motivated contributions from many outside fields. Our approach to studying the immune system computationally has been pragmatic: we have applied any analysis method necessary to understand questions motivated by experimental biology, rather than use biology specifically to discover new physics or methods. Our approach has led us to study problems that span a wide range of time and length scales and require a diverse set of solutions. This thesis describes three projects that span the extremes of this range, from nanometers over nanoseconds to organism-wide responses over hours. The first project was motivated by a puzzle: experimentalists had reached opposing conclusions on the role of a peptide fragment in the main protein interaction responsible for immune recognition. We used molecular dynamics simulations of the proteins to resolve the contradiction by creating a unifying model. The second and third projects jump from the molecular to the system-wide. In the second project, we sought to understand which phenotypes of cancer-fighting immune cells were the most important. To do this, we developed novel data visualizations and applied multivariate dimensionality reduction and regression to understand high-dimensional immunotyping data collected on the phenotypes. The final project addressed an important question in immunology: how accurately do blood assay results reflect the immune response in the tissues, where it matters most? We explored this relationship using a supervised learning model of a highly multidimensional dataset that combined blood and tissue measurements. We found that the two environments could be drastically different and that the relationship mapping blood to tissue was complex. Combined, these three projects highlight the variety of scientific questions and richness of insight that occur at the intersection of immunology and computation. Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemistry, 2012.; This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.; Cataloged from student-submitted PDF version of thesis.; Includes bibliographical references (p. 139-147). Sun, 01 Jan 2012 05:00:00 GMT http://hdl.handle.net/1721.1/78361 2012-01-01T05:00:00Z http://hdl.handle.net/1721.1/74979 Sol-gel deposition of oxide-ion conducting thin films ; and, Liquid precursors to hafnium and tantalum carbides Pell, Jennifer Woodbridge Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Chemistry, 1999.; Includes bibliographical references. Fri, 01 Jan 1999 05:00:00 GMT http://hdl.handle.net/1721.1/74979 1999-01-01T05:00:00Z