Transcription regulation in eukaryotes is a complex process governed by the concerted action of different factors. The work in this thesis is focused on transcriptional regulation in Saccharomyces cerevisiae. I analyzed the regulation of the phospholipid biosynthetic gene INO1 , which has been a model gene for transcription studies for over three decades. Some major questions that I have addressed are: what kinds of cis regulatory sequences and trans factors are important for regulation of INO1? What is the sequence of events in this regulation? How is the recruitment of these trans factors consequential for INO1 transcription? I present my results here for the role of the basic helix loop helix transcription factor (bHLH) family in coordinated regulation of INO1 transcription. I report that the centromeric binding factor 1 (Cbf1p) together with two other members of the bHLH protein family, Ino2p and Ino4p, are required for efficient derepression of INO1 transcription. Together these bHLH transcription factors recruit the ISW2 chromatin-remodeling complex onto the INO1 promoter to drive productive transcription from the INO1 locus. My efforts in studying the regulation of INO1 led me to study the regulation of SNA3, a gene found in tandem upstream (→→) to the INO1 gene and regulated by the same environmental conditions as INO1. Studies on the mechanism of coregulation of adjacent genes in budding yeast have been largely speculative. I provide evidence that the same bHLH proteins which regulate INO1 also regulate SNA3, albeit differentially. Significantly, my results also show that the regulation of both SNA3 and INO1 is dictated from the intergenic region between the two genes. This is a novel mechanism of transcription regulation in yeast as regulation from downstream of ORF is unknown in yeast. Thus, my results with both SNA3 and INO1 provide novel details on how the process of transcription is regulated in response to an environmental cue.

The Drosophila reaper, head involution defective ( hid), and grim genes play key roles in regulating the activation of programmed cell death. In this first part of this study we use the Gal4/UAS targeted gene expression system to demonstrate that unlike reaper or hid, expression of grim alone is sufficient to induce ectopic embryonic CNS midline cell death. We also show that in both the midline and adult eye, grim-induced cell death is not blocked by the Drosophila anti-apoptosis protein Diap2, which does block both reaper and hid-induced cell death. grim can also function synergistically with reaper or hid to induce midline cell death. Finally we analyzed the function of a truncated Reaper-C protein which lacks the NH2-terminal 14 amino acids that are conserved between Reaper, Hid, and Grim. Ectopic expression of Reaper-C revealed cell killing activities distinct from full length Reaper, and indicated that the conserved NH2-terminal domain acts in part to modulate Reaper activity. In the second part of this study, we have analyzed the importance of the RHG motifs in Reaper and Grim for their different abilities to activate cell death during development. Analysis of chimeric R/Grim and G/Reaper proteins indicated that the Reaper and Grim RHG motifs are functionally distinct and help to determine specific cell death activation properties. A truncated Grim-C protein lacking the RHG motif retained an ability to induce cell death and unlike Grim, R/Grim, or G/Reaper, it's actions were not efficiently blocked by the cell death inhibitors Diap1, Diap2, P35, or a dominant/negative Dronc caspase. Finally, we identified a second region of sequence similarity in Reaper, Hid, and Grim, that may be important for shared RHG motif-independent activities. In the third part of this study, we describe a genetic modifier screen that was performed in an attempt to identify other genes that act in cell death pathways. Three proteins from the Ubiquitin/Proteasome (Ub/Pro) pathway were identified as enhancers of the eye cell death phenotype induced by a Reaper/Grim chimeric protein. In particular, a gene encoding a novel F-box protein, Morgue, has been identified.

Polyhydroxyalkanoates (PHAs) comprise a family of macromolecules produced by many bacteria as a carbon and energy reserve and are perceived to have commercial potential as biodegradable thermoplastics. To investigate the biogenesis of PHA inclusion-bodies and the functions of inclusion-body proteins from Bacillus megaterium strain 11561, we identified and cloned a 7.9 kb DNA fragment harboring five genes, phaP, -Q, -R, -B, and -C specifying proteins having known or putative functions in PHA metabolism and/or inclusion-body biogenesis. Sequence similarities to known pha genes identified phaB and -C as specifying acetoacetyl-CoA reductase and PHA synthase, respectively. Putative proteins encoded by phaP, -Q, and -R were not ascribed functions due to lack of significant similarities to known proteins. Both the functionality of the pha gene cluster with respect to PHA accumulation and the transcriptional organization of the genes were determined. Subsequent studies were carried out to further investigate functions of phaP, -Q, and -R. PhaP was established as a major PHA inclusion-body associated protein and was shown to localize to inclusion-bodies in living cells. Further, we demonstrated a phasin-like role for this protein due to its affect on the formation of PHA inclusion-bodies. In addition, our data is consistent with PhaP functioning as a storage protein, implying that the role of PHA inclusion-bodies may be that of a reserve of amino acids in addition to reduced carbon. Regulation of phaP was influenced by PhaQ We showed that PhaQ is a transcriptional repressor of phaP. Moreover, we demonstrated the binding of PhaQ to inclusion-bodies, suggesting that its mode of regulation may involve its localization. Similarly, we showed that PhaR is bound to PHA inclusion-bodies. Our data demonstrated the requirement of phaR for PHA accumulation in vivo and that both PhaC and PhaR were necessary for PHA synthase activity in vitro. Evidence suggests that PHA synthase from strain 11561 can exist in an active or inactive state and that this state is either directly or indirectly influenced by PhaR. A working model is proposed to describe the roles of PhaP, -Q and -R in the metabolism of PHA and biogenesis of PHA inclusion-bodies.

Eukaryotic cells contain scores of small nucleolar RNAs (snoRNAs) that are complexed with proteins in particles known as snoRNPs. SnoRNPs function in processing of ribosomal RNA (rRNA) and nucleotide modification of ribosomal and other RNAs that pass through the nucleolus. The modifications involved are: (1) isomerization of uridine to pseudouridine (Ψ), and; (2) methylation of the 2′-hydroxyl of ribose moieties (Nm). The Ψ and Nm modified nucleotides are the most abundant in rRNA. The purpose of the thesis work is to obtain a better understanding of the synthesis of snoRNPs and the function(s) of the modifications they form. To this end, three lines of research were carried out. Two focus on different snoRNA-associated proteins. The third examines the effect of snoRNP-mediated pseudouridylation of rRNA on ribosome structure and function. One of the snoRNA-associated proteins studied is a conserved helicase previously isolated from a mouse nuclear extract on the basis of association with a conserved snoRNA structure known as the box C/D motif. Here, it was shown that the yeast ortholog of this protein (Rvb2p) is required for production of snoRNAs (both major families) and for localization of snoRNP proteins. The findings are consistent with a role for Rvb2p in the assembly or trafficking of snoRNPs. A second project was aimed at determining if a particular H/ACA snoRNP protein is the enzyme that forms Ψ in rRNA. The protein, Cbf5p, was previously shown to have limited sequence homology with known Ψ synthases (96). Here, in collaboration with the John Carbon laboratory, it was determined that point mutations in Cbf5p in this region of homology can abolish in vivo pseudouridylation of rRNA, arguing that Cbf5p is indeed the global Ψ synthase. Finally, the function of Ψ in the peptidyl transferase center (PTC) was investigated by depleting cells of snoRNPs that modify this domain, and evaluating the under-modified ribosomes. Interestingly, cells lacking Ψs in the PTC are impaired in growth and protein synthesis. Chemical probing revealed changes in the structure of the mutant ribosomes. Enhanced reactivity with dimethylsulfate was observed for nucleotides predicted to influence binding of elongation and initiation factors. The results demonstrate that the structure and activity of yeast ribosomes depend on pseudouridylation of the PTC. The possibility that guide snoRNPs have function(s) other than modification is discussed.

Despite current advances in understanding breast cancer, in 2007 an estimated 178,480 new cases were diagnosed in female patients in the United States, and an estimated 40,640 deaths will occur due to this disease [1]. Numerous risk factors exist for breast cancer including current age, reproductive events, exposure to exogenous hormones or ionizing radiation, and genetic factors [2]. The research presented here is designed to answer questions specifically related to pregnancy and breast cancer risk (parity-induced protection against breast cancer), as well as the molecular pathways involved. We first examined the relationship between p53, a known tumor suppressor gene of breast carcinoma in the mammary gland, and age at first parity in human breast tissue. Using explant cultures from reduction mammoplasty patients, we demonstrated an increase in radiation-induced accumulation of p53 in breast tissue from patients who have undergone an early parity compared to nulliparous patients, or late parous patients. We also demonstrated that p53 accumulation was positively correlated with an increasing number of pregnancies. These are the first studies using cultured human mammary tissue that demonstrate results consistent with experimental rodent models and observational human data. Next, a potential tumor suppressor gene within the breast, cellular retinol binding protein-1 (CRBP1), was found to be significantly up-regulated in response to parity in mice and humans, and that it utilizes transforming growth factor-beta (TGF-β) to confer its signals. Finally, we demonstrated that addition of insulin-like growth factor-I (IGF-I) to mouse mammary gland whole organ cultures (mWOCs) treated with estrogen and progesterone (E+P) could block the normal induction of apoptosis, p53, and its downstream target p21 in response to gamma radiation. In addition, while treatment with E+P resulted in the up-regulation of insulin-like growth factor binding protein-3 (IGFPB3), TGF-β1, and CRBP1, this up-regulation was blocked upon co-treatment with IGF-I. Overall, the research presented here has furthered our knowledge of possible mechanisms by which an early parity can protect against breast cancer. These data suggest that numerous mechanisms can aberrantly alter signaling pathways involved in this protection such that the balance is shifted away from tumor suppression and growth arrest, and towards a phenotype that demonstrates an increased susceptibility toward cancer.

Trypanosoma brucei and related parasites are causative agents of severe diseases that affect global health and economy. T. brucei is responsible for sleeping sickness in humans (African trypanosomiasis) and a wasting disease in livestock. More than 100 years after T. brucei was identified as the etiological agent for sleeping sickness, available treatments remain inadequate, complicated by toxicity, lengthy and expensive administration regiments, and drug-resistance. There is clear need for the development of a new antitrypanosomal drugs. Due to the unique evolutionary position of these early diverging eukaryotes, trypanosomes posses a number of biological properties unparalleled in other organisms, including humans, which could prove valuable for new drug targets. One of the most distinctive properties of trypanosomes is their mitochondrial DNA, called kinetoplast DNA (kDNA). kDNA is composed of over five thousand circular DNA molecules (minicircles and maxicircles) catenated into a topologically complex network. Replication of kDNA requires an elaborate topoisomerase-mediated release and reattachment mechanism for minicircle theta structure replication and at least five DNA polymerases. Three of these (POLIB, POLIC, and POLID) are related to bacterial DNA polymerase I and are required for kDNA maintenance and growth. Each polymerase appears to make a specialized contribution to kDNA replication. The research described in this dissertation is a significant contribution to the field of kDNA replication and the advancement of kDNA replication proteins as putative drug targets for sleeping sickness. Functional characterization of POLIB indicated that it participates in minicircle replication but is likely not the only polymerase contributing to this process. Gene silencing of POLIB partially blocked minicircle replication and led to the production of a previously unidentified free minicircle species, fraction U. Characterization of fraction U confirmed its identity as a population of dimeric minicircles with non-uniform linking numbers. Fraction U was not produced in response to silencing numerous other previously studied kDNA replication proteins but, as we demonstrated here, is also produced in response to POLID silencing. This common phenotype led us to hypothesize that POLIB and POLID both participate in minicircle replication. Simultaneously silencing both polymerases completely blocked minicircle replication, supporting a model of minicircle replication requiring both POLIB and POLID. Finally, we demonstrate that disease-causing trypanosomes require kDNA and the kDNA replication proteins POLIB, POLIC, and POLID. These data provide novel insights into the fascinating mechanism of kDNA replication and support the pursuit of kDNA replication proteins as novel drug targets for combating African trypanosomiasis.

This thesis is the first large-scale molecular phylogenetic study on filarial parasites (family Onchocercidae) which includes 16 species of 6 genera: Brugia malayi, B. pahangi, B. timori, B. patei, B. beaveri, B. buckleyi, Wuchereria bancrofti, W. kalimantani, Mansonella perstans, Loa loa, Onchocerca volvulus, O. ochengi, O. gutturosa, Dirofilaria immitis, Acanthocheilonema viteae and Litomosoides sigmodontis. Two sets of sequence data (5S rDNA spacer and HHa I repeat) were collected by PCR, cloning and dideoxy sequencing. The 5S rDNA gene spacer region sequences were aligned and analyzed by maximum parsimony algorithms, distance methods (neighbor-joining, UPGMA and Fitch-Margoliash methods) and maximum likelihood methods to construct phylogenetic trees. Bootstrap analysis was used to test the robustness of the different phylogenetic reconstructions. The data indicated that both 5S spacer region sequences and Hha I repeat sequences are highly conserved within species yet differ significantly between species. Spliced leader sequences were observed in all of the 5S rDNA spacers sequenced with no sequence variation, although flanking region length heterogeneity was observed even within species. The 5S rDNA spacer sequence data set was used to resolve the branching order of the six genera studied. The various tree-building methods gave very similar results. This study identified four clades which are strongly supported by bootstrap analysis: the Brugia clade; the Wuchereria clade; the Brugia-Wuchereria clade and the Onchocerca clade. Ascaris lumbricoides was used as an outgroup to root the trees generated using 5S sequence data. The analyses indicated that L. sigmodontis and A. viteae may be the most primitive among the 16 species studied. The results of this study do not support the classification which places Loa loa and D. immitis into the Dirofilariinae subfamily. Hha I repeat sequences are 322 nucleotides long, highly repeated and tandemly arranged sequences unique to the nuclear genomes of the genus Brugia. Phylogenetic analyses on the Hha I repeat sequence data set identified at least two clades: the B. pahangi-B. beaveri clade and the B. malayi-B.timori-B. buckleyi clade. Results from the analyses of both data sets were combined and a hypothesis for the phylogeny of the sixteen species of filarial parasites was proposed.

We have used in vivo genomic footprinting to investigate the protein-DNA interactions within the conserved DNA elements (CDE I, CDE II, and CDE III) in the centromere from chromosome III of the yeast Saccharomyces cerevisiae. The in vivo footprint pattern obtained from wildtype cells shows that some guanines within the centromere DNA are protected from methylation by dimethylsulfate (DMS). These results are consistent with studies demonstrating that yeast contain sequence-specific centromere DNA binding proteins. Our in vivo experiments on chromosomes with mutant centromeres show that some mutations which affect chromosome segregation also alter the footprint pattern caused by proteins bound to the centromere DNA. The results of this study provide the first fine-structure map of proteins bound to centromere DNA in living yeast cells and suggest a direct correlation between these protein-DNA interactions and centromere function. We have identified a putative Saccharomyces cerevisiae homologue of a phosphoinositide-specific phospholipase C (PI-PLC) gene, PLC1, which encodes a protein most similar to the $\delta$-class of PI-PLC enzymes. The PLC1 gene was isolated during a study of yeast strains that exhibit defects in chromosome segregation. plc1-1 cells show an 8-9 fold increase in aberrant chromosome segregation and arrest primarily as large-budded or unbudded cells. Molecular analysis revealed that PLC1 encodes a predicted protein of 94 kD with approximately 50% and 26% identity to the highly conserved X and Y domains of PI-PLC isozymes from human, bovine, rat, and Drosophila melanogaster. The putative yeast protein also contains a consensus EF-hand domain that is predicted to bind calcium. Interestingly, the temperature sensitive and chromosome segregation plc1-1 phenotypes are suppressed by exogenous calcium, suggesting a possible relationship between the PI-PLC signaling pathway and control of nuclear events in yeast.

Preparations of unrestricted DNA from Pseudomonas cepacia obtained by lysis of bacteria in agarose plugs contained small amounts of DNA with three different electrophoretic mobilities. The behavior the different species in CHEF gels corresponded to DNA fragments of 3.4, 2.5, and 0.9 Mb. Further analyses were consistent with the notion that the different species represented populations of linear DNAs derived by random double strand breakage of three distinct circular replicons. Treatment of DNA from strain 17616 with SwaI, PacI, and PmeI, restriction enzymes which recognize 8-bp sequences, generated four, six, and three fragments, respectively. Double digestion of the DNA with SwaI and PacI or with SwaI and PmeI generated fragments whose combined molecular weights was 6.8 Mb. Treatment of the DNA with AflII AseI, DraI, SpeI, and XbaI, restriction enzymes which recognize 6-bp sequences, generated about 50 fragments whose combined molecular weight also was about 6.8 Mb. The results indicate that the size of the genome of strain 17616 was 6.8 Mb. The combined molecular weights of the sets of SwaI, PacI, and PmeI fragments was 4.3, 5.9, and 5.9 Mb, respectively, reflecting the fact that the 2.5 Mb replicon contained no SwaI site and that the 0.9 Mb replicon lacked PacI and PmeI sites. The organization of SwaI, PacI, and PmeI fragments comprising each of the replicons was determined by Southern hybridization experiments using various junction fragments as probes and by further restriction analyses. Southern hybridization experiments using r-RNAs and various IS elements from P. cepacia as probes indicated that each of the replicons contained r-RNA genes and IS elements. A SwaI site was introduced into Tn5-751 and the resulting transposon was used to generate and map mutations related to amino acid biosynthesis and the degradation of various carbon sources. Derivatives of strain 17616 were identified in which large segments of the 2.5 and 0.9 Mb replicons had been deleted. The results provide a framework for further genetic analysis of P. cepacia.

<p> Jun N-terminal kinase (JNK) signaling is a key modulator of development and disease in all multicellular organisms. One process in which the consequences of both gain and loss of JNK signaling can be monitored is embryonic dorsal closure (DC) in the fruit fly, <i>Drosophila melanogaster.</i> DC occurs midway through embryogenesis; it is the process by which the lateral epidermis expands bilaterally to meet and fuse at the dorsal midline, thereby encasing the entire embryo in epidermis. JNK signaling in leading edge (LE) cells (the dorsal-most row of epidermis) initiates closure. My studies of a novel but conserved JNK signaling antagonist, Raw, have provided several unique insights into: 1) Jun function as a component of the AP-1 transcription factor, and 2) the role of the epidermis as a signaling template mediating development of the epidermis and adjacent tissues.</p><p> My graduate work has built upon the demonstration that <i>raw</i> is required to prevent promiscuous JNK signaling in the embryonic epidermis just prior to DC. I have shown that <i>raw</i> is necessary for proper accumulation of Jun in LE cells required to define the LE, which functions as a signaling center required for epidermal closure as well as for underlying heart development. I have gone on to show that Jun accumulates at previously unrecognized sites in the embryonic epidermis, including tracheal pits and solitary epidermal cells lying directly above the peripheral nervous system (PNS). Jun activity is required for tracheal and nervous system defects observed in mutants of two JNK signaling antagonists, <i>raw</i> and <i> rib,</i> and indicates that cell signals within and to an adjacent tissue are integral to proper development. I have found that the epidermis plays an instructive role during development, and results from my work have led to insights into how JNK signaling centers in the epidermis coordinate morphological processes.</p><p> As Raw is a novel but conserved JNK signaling antagonist, I have built and tested models of its molecular mechanism of action as well. Bolstering conclusions of previous studies of mammalian c-Jun in cell culture, my data indicate that N-terminal phosphorylation is not an on/off switch, but rather it increases Jun stability for its activity as a component of the AP-1 transcription factor. <i>raw</i> mutants exhibit both high levels of Jun protein and an accumulation of phosphorylated Jun (P-Jun), and my data point to a role for Raw in effecting the Jun:P-Jun ratio via mediation of Jun degradation. In deciphering the mechanism of Raw function, we are gaining significant new insights into previously unrecognized mechanisms of JNK signaling regulation. Understanding these mechanisms will be important for dissecting the etiology of developmental abnormalities and diseases, such as cancer, which hinge on the Goldilocks effect, having just the right amount of signaling at just the right time.</p>