acase@tulane.edu / Next generation sequencing (NGS) is a relatively new technology that has revolutionized the way scientists discover and investigate pathogens. It has been estimated that a staggering one in every five cancers worldwide is linked to an infectious agent. An understanding of the pathogen biology as well as the interactions with the host will lead to better therapies and outcomes for patients suffering from pathogen-associated malignancies. Despite the promise for this phenomenon through NGS-based approaches, we are still in the infancy of sequence analysis and are unable to fully appreciate the potential of NGS. To facilitate data mining, an automated computational pipeline for the simultaneous analysis of pathogen and host transcripts called RNA CoMPASS was developed. Using RNA CoMPASS to investigate a variety of sequencing datasets over the years, substantial bacterial contamination have been routinely identified in human-derived RNA-seq datasets that likely arose from environmental sources. Based on this analysis, a need for more stringent sequencing and analysis protocols to investigate sequence-based microbial signatures in clinical samples is crucial. NGS-based approaches were utilized to investigate the role of Epstein-Barr virus (EBV) in the pathogenesis of gastric carcinoma. A comprehensive assessment of the virome of various brain tissue samples was also performed, with the notion that an NGS-based detection method would be unbiased, sensitive, specific, and accurate. Taken together, these studies provide a framework for using NGS technology to study oncogenic pathogens and bring awareness to contamination issues within sequencing datasets. / 1 / Michael J Strong

Hox genes are regulatory genes encoding small proteins containing a highly conserved 61-amino acid motif, the homeodomain, that enables Hox proteins to bind to DNA at specifically recognised binding sites and transcriptionally activate their target genes. In mammalian species there are 39 Hox genes and they are structural and functional homologs of the Drosophila homeotic complex (Horn-C). During embryogenesis and early development the Hox genes are expressed in a spatiotemporal fashion, where they operate as master transcriptional regulators. Hox genes are further expressed in fully differentiated adult cells, potentially in a tissue-specific manner involving maintenance of the normal phenotype. In selected oncogenic transformations, dysregulated Hox gene expression has been observed, indicating an involvement of these transcriptional regulators in carcinogenesis and metastasis. Utilising quantitative real-time PCR assays, these studies investigated the expression patterns of 20 Hox genes and two wellcharacterised Hox cofactors (Pbx and Meis) in malignant and non-malignant human breast and skin cancer cells. Dysregulated Hox expression was observed for all malignancies tested, of which some misexpressed Hox genes seemed random, whereas other Hox transcripts showed altered levels potentially corresponding with the invasive capacity of the cells. Also, the Hox cofactors Pbx and Meis showed no marked changes in expression levels from the non-malignant to the malignant phenotypes, indicating that it is dysregulated Hox gene expression rather than dysregulated gene expression of Hox cofactors that potentially commit the cell to redifferentiate and undergo oncogenic transformation. Although the Hox proteins are known to be key transcriptional regulators of development, the mechanisms by which they gain their in vivo functional specificity is still largely unknown. They all show strikingly similar transcriptional specificity in vitro, yet show unique specificity in their in vivo environment. This paradox has been the subject of intense scrutiny, however very few direct Hox target genes have been identified, making it a difficult task to decipher the exact manner in which Hox proteins exert their functional potential. Therefore, the studies presented herein were aimed at identifying further Hox target genes in the human system. Utilising differential display approaches, several potential downstream target genes were isolated. Substantiated with real-time PCR assays, one of these potential targets was selected as a likely direct Hox gene target, and as such subjected to further studies. By the combination of bioinformatic analyses, transfection protocols and luciferase assays, a gene encoding the SR-related protein SRrpl3O was shown to be trans-activated in vitro by HOXD4 via a putative Hox binding element within its promoter region. This is the first reported link between Hox transcription factors and the SR and SR-related family of pre-mRNA splicing proteins, offering a new and exciting insight into the complex nature of Hox functional specificity. Finally, this thesis also puts forward new ideas regarding how the Hox proteins gain their transcriptional and functional specificity. Utilising bioinformatic tools in conjunction with performing an extensive review of the disparate catalogue of Hox-related research reports, work herein offers the first comprehensive analysis of the mammalian Hox gene targets in relation to their promoter structures, as well as with respect to the expanded Hox DNA-binding elements. This work reports that identified Hox targets generally contain TATA-less core promoters, many of which have several GC-box elements. The Hox binding elements show no apparent preference regarding their location relative to the transcription start site (TSS), as they are found both upstream and downstream of the TSS, as well as being located close to proximal core promoter elements for some genes and at more distant positions in other gene promoters. Finally, the core Hox binding element TAAT/ATTA contains only part of the necessary recognition sequence involved in Hox-DNA binding, and the notion that flanking base pairs dictate trans-regulatory potential is further explored with the hypothesis that the immediate 3' base pair dictates an activator/repressor-switch of the Hox trans-regulatory effect.

We have developed crown-gall resistant transgenic plants capable of suppressing Agrobacterium tumefaciens T-DNA oncogenes. Crown gall tumors result from overproduction of auxin and cytokinin in plant cells transformed by A. tumefaciens. High phytohormone levels result from expression of two auxin biosynthetic genes, tryptophan monooxygenase (iaaM) and indole acetamide hydrolase (iaaH), and isopentenyl transferase (ipt), which mediates cytokinin synthesis. Inactivation of ipt and either one of the two auxin biosynthesis genes prevents crown gall formation. To suppress T-DNA oncogene expression, we created transgenic tobacco that produce the corresponding untranslatable sense-strand RNAs. This phenomenon, called cosuppression, frequently blocks expression of transgenes in plants. Often, expression of an untranslatable sense-strand transgene elicits sequence-specific destruction of both the mutant mRNA and the corresponding wild-type mRNA. Here we show that cosuppression can block expression of A. tumefaciens T-DNA oncogenes, resulting in plants that are resistant to gall induction by certain strains of A. tumefaciens. / Graduation date: 1999

published_or_final_version / Microbiology / Master / Master of Philosophy

The effects of dibutyryl 3'; 5' cyclic monophosphate (dbcAMP) on several parameters of transformation were studied using a Kirsten sarcoma virus (Ki-MSV)-transformed mouse cell line (K-A31). Treated cells showed changes in morphology, decreased motility, saturation density and growth rate, and lost the capacity for anchorage-independent growth. In contrast to many other transformed cell lines, fibronectin and an elaborate cytoskeleton were present in K-A31 cells. The transcription of the proviral genome was examined using both reverse-transcribed and nick-translated ('3)H-DNA probes, and certain viral-specific RNAs were found restricted to the nucleus of dbcAMP-treated cells. Additive hybridization experiments suggested these RNAs were transcribed from rat-derived sequences located in the 5'-half of the proviral genome. These results are discussed with respect to the properties most closely associated with cellular malignancy, and the possible mechanism of dbcAMP-mediated reverse-transformation in K-A31 cells.

Thesis (Ph.D.) -- McMaster University, 1997. / Includes bibliographical references (leaves 204-216). Also available via World Wide Web.

Thesis (Ph. D.)--University of Missouri--Columbia, 1998. / "May 1998." Typescript. Vita. Includes bibliographical references (leaves 172-189). Also available on the Internet.

Bone marrow stromal antigen 2 (BST-2/tetherin) is a type II transmembrane protein that plays various roles, including protective and detrimental roles in the host. Cellular responses to BST-2 expression or the lack thereof, may be cell type and context-dependent and may vary with time. When protective, BST-2 functions as an antiviral factor, renowned for its ability to tether budding enveloped viruses to the membrane of infected cells. Tethering of budding virions prevents their release into the extracellular milieu limiting infection of naïve cells. The antiviral role of BST-2 has been predominantly studied using cultured cells. Insight into the role of BST-2 in inhibition of viral infection in vivo came from our study of the alphavirus Chikungunya virus (CHIKV) and the retrovirus mouse mammary tumor virus, (MMTV). BST-2 prevents the release of CHIKV and MMTV virions from infected cells and limits the replication of both viruses in mice. In the context of CHIKV infection, BST-2 protects the host in a tissue-type dependent manner. In lymphoid and most non-lymphoid tissues, expression of BST-2 limits CHIKV replication. In addition, BST-2 regulates CHIKV-induced inflammatory responses in mice, an indication that BST-2 may function to initiate and amplify innate immune responses. Host response to MMTV infection depends on the stage of the infection and disease sequela. Acute infection of immune cells with MMTV results in an initial increase in BST-2 expression followed by a sharp decline. In contrast, in MMTV-induced mammary tumors, BST-2 mRNA and protein are elevated, so is the viral load. This is an indication that the antiviral role of BST-2 is not operative once mammary tumors have developed. These data provided the initial evidence that BST-2 may promote breast cancer progression. Indeed, data from two mouse models of breast cancer show that expression of BST-2 is necessary for cell to cell and cell to extracellular matrix interactions. Thus, BST-2 expression in breast cancer cells enhances cancer cell adhesion, anchorage-independency, migration, and invasion, culminating in increased tumor mass, increased metastases, and reduced host survival. Structurally, BST-2 homodimerization is important for its cancer-promoting role as dimers of BST-2 regulate anchorage-independency, resistance to anoikis, and enhanced adhesion between cancer cells and components (proteins and cells) of the tumor microenvironment. How BST-2 is enriched in breast cancer cells was elusive until our in silico analyses of a large human breast cancer dataset that revealed the involvement of epigenetic regulation of BST-2 in breast tumors. In highly aggressive breast cancers, specific CpG sites in and at close proximity to the BST-2 promoter are hypomethylated. This is in sharp contrast to non-aggressive luminal cancers and normal breast epithelial cells. These data suggest that a progressive loss of methylation on the BST-2 gene may contribute to constitutive overexpression of BST-2 in tumors. Overall, these findings show that 1) BST-2 contributes to the emergence and progression of breast malignancies and may be used as a therapeutic target or as a biomarker for aggressive breast cancers; and, 2) BST-2 acts as a viral sensor to initiate antiviral inflammatory responses and could be exploited therapeutically to treat viral infections. We highlight the need for additional research on the antiviral and cancer-promoting roles of BST-2 to reconcile both functions for the purpose of therapeutics.

No description available.

No description available.