Thesis (Ph. D.)--Northeastern University, 2008. / Title from title page (viewed March 3, 2009). Graduate School of Arts and Sciences, Dept. of Physics. Includes bibliographical references (p. 133-155).

Thesis (M.S.)--Brandeis University, 2009. / Title from PDF title page (viewed on May 29, 2009). Includes bibliographical references.

Thesis (M.S.)--University of Alabama at Birmingham, 2008. / Title from PDF title page (viewed Sept. 21, 2009). Additional advisors: Martha W. Bidez, W. Michael Carson, Richard A. Gray, W. William Wilson. Includes bibliographical references.

Indiana University-Purdue University Indianapolis (IUPUI) / In an effort to advance the methodology in analyzing RNAPII protein-protein interaction network and to determine the role of the CTD in controlling RNAPII transcription, we devised a method to specifically isolate the CTD-associated and CTD-less RNAPII to identify the proteins that interact with both the CTD and the globular core of RNAPII using novel purification scheme coupled to quantitative proteomics.

Bacteria live under constant assault by bacteriophages and have evolved a diverse array of defense strategies. CRISPR-Cas systems are prokaryotic adaptive immune systems that rely on RNA-guided binding for the recognition and degradation of invading nucleic acids. Intriguingly, some bacteria also encode divergent CRISPR-Cas systems that can bind to — but cannot degrade — target nucleic acids. In this dissertation, I describe the study of nuclease-deficient CRISPR-Cas systems alongside the evolutionary pressures that led to their persistence in bacterial genomes. I present experimental data for the existence of CRISPR-associated transposons (CASTs) that utilize the RNA-guided DNA binding ability of Type I-F CRISPR-Cas systems to direct transposition to new target sites in a heterologous Escherichia coli host. This RNA-guided DNA integration pathway can tolerate large cargos of up to 10 kilo-base pairs in size, and is highly specific for the programmed target site, as determined by deep sequencing experiments. We further reveal the physical link between the CRISPR-Cas and transposition machineries through biochemical experiments and by determining cryo-EM structures of the transposition protein TniQ in complex with the CRISPR-Cas effector. After bioinformatic analyses and experimental validation we established an array of twenty diverse CAST systems for which a subset works completely orthogonally. This dataset revealed the modular nature of CASTs by showcasing the horizontal acquisition of targeting modules and by characterizing a system that encodes both a programmable, RNA-dependent pathway, and a fixed, RNA-independent pathway. Further analysis of the transposon-encoded cargo genes uncovered the striking presence of anti-phage defense systems, suggesting a role in transmitting innate immunity between bacteria. Finally, we exploit high-throughput screening assays to determine the specific sequence and spacing requirements of the transposon ends, and use this knowledge to develop a CAST-mediated endogenous gene-tagging approach. Intriguingly, our experiments uncover the involvement of a previously unknown cellular protein, integration host factor (IHF), which is critical for transposition of VchCAST, but not other homologous systems. Collectively, the work presented in this dissertation describes the discovery of RNA-guided DNA integration employed by CASTs, substantially advances our biological understanding of these systems, and expands the suite of RNA-guided transposases for programmable, large-scale genome engineering.

Master of Science / Department of Computing and Information Sciences / Doina Caragea / William H. Hsu / The term 'protein-protein interaction (PPI)' refers to the study of associations between proteins as manifested through biochemical processes such as formation of structures, signal transduction, transport, and phosphorylation. PPI play an important role in the study of biological processes. Many PPI have been discovered over the years and several databases have been created to store the information about these interactions. von Mering (2002) states that about 80,000 interactions between yeast proteins are currently available from various high-throughput interaction detection methods. Determining PPI using high-throughput methods is not only expensive and time-consuming, but also generates a high number of false positives and false negatives. Therefore, there is a need for computational approaches that can help in the process of identifying real protein interactions. Several methods have been designed to address the task of predicting protein-protein interactions using machine learning. Most of them use features extracted from protein sequences (e.g., amino acids composition) or associated with protein sequences directly (e.g., GO annotation). Others use relational and structural features extracted from the PPI network, along with the features related to the protein sequence. When using the PPI network to design features, several node and topological features can be extracted directly from the associated graph. In this thesis, important graph features of a protein interaction network that help in predicting protein interactions are identified. Two previously published datasets are used in this study. A third dataset has been created by combining three PPI databases. Several classifiers are applied on the graph attributes extracted from protein interaction networks of these three datasets. A detailed study has been performed in this present work to determine if graph attributes extracted from a protein interaction network are more predictive than biological features of protein interactions. The results indicate that the performance criteria (such as Sensitivity, Specificity and AUC score) improve when graph features are combined with biological features.

The oncoprotein MDM2 has an integral role in cancer development via multiple signalling pathways. Two proteomic mass spectrometry screens, label-free with spectral counting quantitation and 8-plex iTRAQ were used to identify proteins up or downregulated over time by the MDM2 targeting drug Nutlin. A subset of previously identified MDM2 binding partners were identified as altered after Nutlin treatment, along with proteins which have not as yet been linked to MDM2 or p53. Proteins altered two hours after Nutlin treatment were screened for sequence similarity to an MDM2 binding consensus motif based on the BOX-I region of p53. Peptides corresponding to this motif were validated for MDM2 binding, and the mode of binding investigated using competition ELISA and thermal denaturation assays. Known MDM2 ligands such as Nutlin were shown to have a range of effects on the binding of these newly identified MDM2 peptides, which may be attributed to allosteric regulation of MDM2. The effects of Nutlin on two full length proteins identified by the MS screens, CypB and NPM, were confirmed in vivo. In vitro binding of MDM2 to CypB and PK, which contain BOX-I like motifs, was also demonstrated validating proteomic mass spectrometry screens as a method to identify new protein-protein interactions. To further investigate the potential of linear motifs to modulate protein-protein interactions, a peptide aptamer targeting the protein AGR2 was tested for effect on AGR2 and p53 in a cancer cell line.

Human papilloma virus (HPV) is a ubiquitous virus implicated in a growing list of cancers, particularly cervical cancer‐ the second most common cancer among women worldwide. Although persistent infection with high‐risk oncogenic HPVs such as types ‐16 or ‐18 is necessary, additional factors like co‐infection with other viruses can play a role in cancer progression. Protein‐protein interactions play a central role in the infection, survival and proliferation of the virus in the host. Although some interactions of HPV proteins are well characterised, it is essential to discover other key viral interactions to further improve our understanding of the virus and to use this knowledge for the development of newer biomarkers and therapeutics. The aim of this study was to systematically analyse the interactions of HPV‐16 proteins using yeast two‐hybrid (Y2H). To achieve this, a clone collection of the viral proteome was generated by recombinatorial cloning and three independent Y2H screens were performed: (i) Intra‐viral screen to identify interactions among the HPV‐16 proteins; (ii) Inter‐viral screen to identify interactions with proteins of Herpes Simplex Virus (HSV) which is suggested to be a co‐factor; and (iii) Virus‐host screen to identify novel cellular binding partners. The intra‐viral Y2H screen confirmed some of the previously known interactions and also identified binding of the E1 and E7 proteins. Deletion mutagenesis was performed to map the interaction domains to the amino‐terminal 92 amino acids of E1 and carboxy‐terminal CxxC domain of E7. Replication assays suggest a possible repression of E1‐mediated episomal replication by direct binding of E7. The inter‐viral Y2H screen identified interactions of HPV proteins with seventeen HSV‐1 proteins including transcriptional regulator ICP4 and neurovirulance factor ICP34.5. The biological relevance of these interactions in the context of co‐infection is discussed. The virus‐host screen performed against a human cDNA library identified 54 interactions, a subset of which was validated by biochemical pull‐down assays. The functional relevance of an interaction between E7 and a proto‐oncogene spermatogenic leucine zipper protein (SPZ1) was further investigated suggesting a role of SPZ1 in E7‐mediated cell proliferation. The work presented in this thesis identifies several novel interactions of HPV proteins. Future work will involve the in‐depth elucidation of biological relevance of these interactions. In particular, the interactions of E7 with E1 and SPZ1 are of great interest to improve our understanding of the life cycle and pathogenesis of the virus which can be applied for improved strategies of prevention and treatment of malignancies caused by HPV.

The MDM2 and MDM4 oncoproteins are both involved in regulating the tumour suppressor, p53. While the MDM2–p53 interface is structurally and biophysically well characterised, the MDM4-p53 interaction has only recently attracted researchers’ attentions. The goal of this project was to establish structural and chemical ground rules for the disruption of the interactions between the N-terminal domains of MDM2/4 and p53, which is an attractive anticancer strategy. In the current work, successful recombinant production and purification protocols for both the N-terminal domains of MDM2 (i.e. MDM2-N, residues 11-118) and MDM4 (MDM4-N, residues 14-111) have been established, yielding protein in sufficient quantity and quality for analysis using nuclear magnetic resonance spectroscopy (NMR). Two screening strategies were employed to identify small-molecule antagonists of the MDM2-N:p53 interaction. First, a virtual screening exercise identified several compounds that were shown (by NMR) to bind to MDM2-N with μM KDs. Docking studies supported by NMR chemical shift perturbation analysis suggested proposals for binding modes. The results are discussed in relation to the previously reported binding to MDM2-N of well-characterised inhibitors of the MDM2:p53 interaction such as Nutlin-3. Second, a fragment-based library was screened against MDM2-N using TROSY-type NMR spectra to monitor binding. Several hits were identified and the results are discussed with regard to the “druggability” of the MDM2-N p53 interaction. To better understand the p53-binding groove of MDM4-N, multidimensional NMR was used to investigate the structure and backbone dynamics of double-isotopically labelled samples of MDM4-N, both free (i.e. apo-MDM4-N) and in complexes with a p53-derived peptide or Nutlin-3. The apo-MDM4-N is more conformationally dynamic than MDM2, since it contains unstructured regions. These regions appear to become structured upon binding of a ligand. MDM4 appears to bind its ligand through conformational selection and/or an induced fit mechanism involving reorganization of key sub-sites within the binding groove. This study highlighted Abstract differences between Nutlin-3 and peptide binding that suggest the rational design of specific inhibitors of the MDM4:p53 interaction.

published_or_final_version / abstract / Biochemistry / Doctoral / Doctor of Philosophy