In this study, the potencies of 2,3,7,8-tetrachlorodibenzo- p-dioxin (TCDD), 2,3,4,7,8-pentachlorodibenzofuran (PeCDF) and 2,3,7,8-tetrachlorodibenzofuran (TCDF) to induce cytochrome P4501A (CYP1A) were determined. The concentration-dependent effects of TCDD, PeCDF and TCDF on ethoxyresorufin O-deethylase (EROD) activity, CYP1A4 mRNA and CYP1A5 mRNA expression were measured in primary hepatocyte cultures of chicken (Gallus gallus), ring-necked pheasant (Phasianus colchicus), Japanese quail (Corturnix japonica ), and herring gull (Larus argentatus) embryos exposed for 24h. The results demonstrated that PeCDF was as potent as TCDD to induce CYP1A in chicken hepatocytes, but that PeCDF was more potent than TCDD in ring-necked pheasant, Japanese quail and herring gull hepatocytes. Moreover, the chicken was the most sensitive species to CYP1A induction by TCDD or TCDF, but all species were approximately equisensitive to CYP1A induction by PeCDF. Together these results showed that all species do not respond the same way to different DLCs, suggesting that interactions between a DLC and a species are structure- and cell type-dependent. These results raise interesting questions concerning the molecular mechanism of action of DLCs and the toxic equivalency factors that are broadly assigned to all avian species.

The renin-angiotensin system (RAS) has been implicated in the pathophysiology of heart failure, hypertension and diabetic nephropathy. Positron emission tomography (PET), a non-invasive imaging modality, can provide information on cellular function and receptor density. Based on previous structure-activity studies, two 11C-labeled analogues of the clinically used AT 1 receptor antagonist candesartan were developed, [11C]-methyl-candesartan and [11C]TH4, and characterized for binding specificity and selectivity. [11C]Methyl-candesartan was further tested to determine the presence of labeled metabolites and potential for small animal PET. [11C]Methyl-candesartan displayed higher specific binding and selectivity to angiotensin II type 1 (AT1) over angiotensin II type 2 (AT2), Mas (Ang(1-7)), beta-adrenergic and alpha 2-adrenergic receptors in kidney regions in ex vivo studies. Selectivity for AT1 over Mas receptors was not observed for [11C]TH4. Renal binding selectivity over AT2 was confirmed in [11C]methyl-candesartan microPET imaging studies. Metabolite analysis of [11 C]methyl-candesartan detected a labeled metabolite in the rat kidney, although the identity has not been determined. The work presented here supports the potential of [11C]methyl-candesartan for in vivo imaging of AT1 receptors.

The acrosome is a large secretory granule that undergoes exocytosis when receptors on the sperm surface bind ligands found on the oocyte's extracellular matrix. Acrosomal exocytosis resembles stimulated neurotransmitter release in neurons in that it is triggered by a rise in intracellular calcium. In neurons, a core complex composed of the SNARE proteins - syntaxin, SNAP-25 and VAMP is involved in synaptic vesicle fusion. Regulation of this protein complex is accomplished through the action of accessory proteins, including complexin, which is thought to stabilize the core SNARE complex prior to fusion. Recent evidence has revealed that isoforms of the SNARE proteins and their accessory proteins are present in mammalian sperm, where they might mediate the exocytosis of the acrosome. It has been hypothesized that the SNARE complex may be regulated as part of the process of capacitation, a set of physiological changes that occur within the female tract, during which the spermatozoon acquires its fertilizing ability. In this thesis, the hypothesis that complexin is involved in the regulation of the acrosome reaction is examined. To do so, streptolysin-O permeabilized sperm will be used to study the effect of recombinant complexin and of an antibody directed against complexin on the rate of the acrosome reaction, as well as on the sensitivity of this fusion machinery to calcium. Regulation may also occur through posttranslational modification or conformational change in one of the SNARE proteins. Syntaxin 2, which is the isoform present in sperm, showed a shift in apparent molecular weight on Western blots with capacitation. Alkaline carbonate extraction and dephosphorylation were used in an attempt to determine the kind of modification syntaxin 2 is undergoing during capacitation. Together, these data will shed light onto the role of complexin in the regulation of the acrosome reactions as well as on the modifications syntaxin 2 undergoes during capacitation.

During zebrafish embryogenesis, a gradient of the secreted TGFβ ligand Nodal induces expression of mesendodermal genes in an exposure-dependent manner, and also induces expression of the secreted Nodal feedback inhibitor Lefty. The long-range Lefty gradient dampens signaling by the shorter-range Nodal gradient, and together they are required for proper patterning of the germ layers. My research has addressed two fundamental questions about this patterning system: 1) How are the distinct ranges of Nodal and Lefty achieved, and 2) What is the role of Lefty-mediated feedback inhibition in this patterning system? To understand why Nodal and Lefty have different activity ranges, we developed fluorescent fusion proteins and microscopy techniques to measure the distributions, degradation rates, and diffusion coefficients of Nodal and Lefty in living zebrafish embryos. We found that Nodal-GFP forms a shorter-range gradient than Lefty-GFP in vivo, consistent with the shorter activity range of Nodal. The extracellular half-lives of Nodal-Dendra2 and Lefty-Dendra2 are similar, but the diffusion coefficient of Nodal-GFP is much lower than that of Lefty-GFP. Differential diffusivity therefore explains the shorter range of Nodal compared to Lefty. To better understand the function of Lefty during vertebrate embryogenesis, we generated and analyzed zebrafish lefty mutants. Although patterning is highly robust to lefty gene dosage, complete loss of lefty causes upregulation of mesendodermal genes during gastrulation, aberrant cell internalization, and severe patterning defects, consistent with excess Nodal signaling. Mutations in nodal partially suppress patterning defects caused by lefty loss, and lefty mutants can be fully rescued by ectopic expression of lefty-gfp or by treatment with low levels of a Nodal inhibitor drug. Further, lefty expression is highly responsive to Nodal signaling. These results demonstrate that Lefty acts as a fast-responding, global dampener of Nodal signaling that restricts mesendoderm specification.

Sorting Nexin 10 (Snx10) is expressed in osteoclasts and is required for osteoclastic bone resorption in vitro. To study the role of Snx10 in osteoclastic bone resorption and bone homeostasis in vivo, we investigated the expression of Snx10 and created mouse models in which Snx10 was deficient in osteoclasts or globally. Osteoclast-specific Snx10-deficient mice exhibited severe osteopetrosis with abnormal bone micro-architectural parameters in vivo, consistent with the failure of osteoclasts to normally resorb bone. Osteoclast-derived Snx10 deficiency didn’t completely inhibit osteoclast formation, however, the capacity to resorb bone was significantly reduced. Intracellular vesicular transport, ruffled border formation and extracellular acidification were found to be severely impaired due to osteoclast-derived Snx10 deficiency. We also discovered that Snx10 was highly expressed in gastric zymogenic cells, with mutations leading to gastric dysfunction and low calcium solubilization. Global Snx10-deficiency in mice results in a combined phenotype: osteopetrosis (due to osteoclast defect) and rickets (due to gastric dysfunction and low calcium availability, resulting in impaired bone mineralization and hypocalcemia). Osteopetrorickets, the paradoxical association of insufficient mineralization in the context of a positive total body calcium balance, was thought to occur due to failure of the osteoclasts to maintain normal calcium homeostasis. However, osteoclast-specific Snx10 deficiency had no effect on calcium balance, and therefore led to severe osteopetrosis without rickets. Moreover, supplementation with calcium gluconate prevented the rachitic phenotype and rescued the early death in global Snx10-deficient mice, suggesting that this may be a life-saving component of the clinical approach to Snx10-dependent human ARO with hypocalcaemia and/or no improvement after HSCT. We concluded that tissue-specific effects of Snx10 mutation need to be considered in clinical approaches to this disease entity. Reliance solely on hematopoietic stem cell transplantation can leave hypocalcemia uncorrected with sometimes-fatal consequences. To our knowledge, this is the first study to explore the role of Snx10 using the genetically modified mouse model. This study not only uncovered the cellular mechanism by which Snx10 regulates osteoclastic bone resorption but also established an essential role for Snx10 in bone homeostasis and underscore the importance of Snx10-dependent gastric function in calcium homeostasis.

Silent chromatin, or heterochromatin, refers to regions of the genome in which genes are constitutively repressed. These regions are important for regulating developmental genes and for maintaining genome stability, and are epigenetically inherited. In Saccharomyces cerevisiae, subtelomeres and silent mating type loci are assembled into silent chromatin by the Silent Information Regulator (SIR) complex, composed of Sir2, Sir3 and Sir4, which deacetylates histones and spreads along chromatin. Many questions remain regarding the mechanism of Sir protein spreading along chromatin and the mechanism of epigenetic inheritance of silent chromatin domains. It has been hypothesized that the lateral Sir-Sir protein interactions together with Sir-nucleosome interactions cooperatively recruit Sir proteins to spread along chromatin. In my thesis project, I set out to test this cooperativity hypothesis by examining the interaction of Sir proteins with well-defined in vitro reconstituted mono- and di-nucleosomes. Using electrophoretic mobility shift assay (EMSA), I find that Sir3, the main nucleosome-binding component of the SIR complex, associates with nucleosomes cooperatively, involving the dimerization of Sir3 bound to neighboring nucleosomes. I demonstrate that this inter-nucleosomal cooperativity is mediated by the Sir3 C-terminal winged helix (wH) dimerization domain and is further stabilized by the Sir4 coiled-coil (CC) domain, which mediates both Sir4 homodimerization and Sir3-Sir4 interactions. There is functional redundancy from the two domains in mediating binding cooperativity, as suggested by the measurement of cooperativity free energy. Surprisingly, my binding measurements suggest that there are no Sir-Sir protein interactions on the same nucleosome. Moreover, by using an in vitro bridging assay, I show that Sir3 effectively bridges free nucleosomes in solution and that its wH domain is required for its bridging activity. My in vitro results are corroborated by in vivo ChIP-seq results showing that either the Sir3 wH domain or the Sir4 CC domain alone could mediate weak spreading of Sir3 protein, away from recruitment sites, under Sir3 overexpression conditions. However, mutations in both domains abolish the spreading completely. Both histone H4 lysine 16 (H4K16) acetylation and histone H3 lysine 79 (H3K79) methylation are hallmarks of euchromatin in S. cerevisiae. I quantify the effect of either modification alone and both modifications in combination on Sir3-nucleosome binding affinity. This shows that either modification alone decreases Sir3 binding affinity towards nucleosomes by 3-4 fold, and that the two modifications work together to reduce the binding affinity even further. Statistical mechanical modeling of the nucleosome binding results indicate that the combined effect of H4K16 acetylation and H3K79 methylation can account for partitioning of Sir3 between silent and active chromatin regions in vivo. Our findings and their quantitative analysis suggest that SIR complexes spread along chromatin discontinuously, arguing against the stepwise polymerization model for silent chromatin assembly. / Medical Sciences

Genes encoding subunits of the SWI/SNF chromatin remodeling complex are frequently mutated in a wide variety of cancers, but the mechanism of cancer formation following SWI/SNF subunit mutation is not known. As a result, improved understanding of how mutation of these subunits promotes oncogenesis could have broad relevance for human cancer therapy. In this dissertation, two distinct lines of investigation were pursued with the common goals of understanding the mechanisms by which SWI/SNF subunit mutation contributes to cancer and identifying vulnerabilities specific to SWI/SNF-mutant cancers. First, a genome-wide screen was used to find specific dependencies conferred by ARID1A mutation, the most frequently mutated subunit of the SWI/SNF complex in cancer. ARID1B was identified as the number one gene specifically essential for the growth of ARID1A-mutant cancers, and ARID1B loss was found to destabilize SWI/SNF and impair cell proliferation in ARID1A-mutant backgrounds. ARID1A and ARID1B were also found to be frequently co-mutated in cancer cell lines and primary samples. Furthermore, these proteins were found to share overlapping roles but also retain unique functions in a cell context dependent manner. Second, DNA methylation was studied as a putative epigenetic mechanism by which mutation of SMARCB1, a core SWI/SNF subunit, leads to genomically stable rhabdoid tumors. Reintroduction of SMARCB1 in rhabdoid tumor cell lines resulted in a decrease of DNA methylation at a subset of upregulated enhancers, identifying DNA methylation as a secondary change following SWI/SNF subunit perturbation. The results from these two distinct lines of study reveal novel vulnerabilities of cancers harboring SWI/SNF mutation and elucidate partial mechanism by which mutation of two unique SWI/SNF subunits contributes to cancer. These findings provide new insight into potential therapeutic opportunities for the wide spectrum of SWI/SNF-mutant cancers. / Medical Sciences

Medium chain fatty acids (MCFAs, 6-12 carbons) are potential precursors to biofuels with properties similar to gasoline and diesel fuel but are not native products of Escherichia coli fatty acid synthesis. Herein we engineer E. coli to produce, metabolize, and activate MCFAs for their future reduction into alcohols and alkanes (potential biofuels). We develop an E. coli strain with an octanoate (8-carbon MCFA) producing enzyme (a thioesterase), metabolic knockouts, and the capability to inducibly degrade an essential metabolic enzyme that would otherwise divert carbon flux away from octanoate. We show that this strain can produce octanoate at 12% theoretical yield. To determine limitations on octanoate catabolism that could prevent its conversion into an acyl-CoA thioester activated for later reduction into alcohols and alkanes, we evolve E. coli to grow on octanoic acid as sole carbon source. We show that our fastest growing evolved strain contains mutations that enhance the expression of acyl-CoA synthetase FadD. We then directly mutate the fadD gene and screen for mutations that enhance growth rate on octanoic acid. In-vitro assays show that the mutations we identify increase FadD activity on MCFAs. These results, homology modeling, and further mutagenesis lead us to hypothesize that our mutations enhance FadD activity by aiding product exit. This work develops a technique (inducible degradation of an essential metabolic enzyme) and generates fadD mutants that should be useful for the production of medium chain biofuels and other compounds. / Medical Sciences

MicroRNAs (miRNAs) and other classes of short non-coding RNAs regulate essential processes in the development and function of the nervous system. Regulation of miRNAs by neural activity has also been reported. Recently, instances of piwi interacting RNA (piRNA) and endogenous short interfering RNA (esiRNA) mediated modulation of neural physiology have been reported. To better understand the role of miRNAs and other classes of short non-coding RNAs in long term memory (LTM) formation, we have conducted high throughput sequencing on 15-35nt RNAs isolated from heads of Drosophila that have been subjected to aversive olfactory conditioning. We developed genome wide profiles of miRNA, piRNA, and esiRNA, and tested for differential expression following conditioning. We find that 5 miRNAs exhibit significant regulation in the conditioned group. We identify several esiRNA generating loci within genes required for olfactory LTM formation. Our data reveal that an intron of the multiple wing hairs (mwh) gene forms secondary structures and generates esiRNAs following conditioning from regions that correspond to lysozyme family genes located within the mwh intron. We find that piRNAs are produced in fly heads, and that a small set of piRNA generating loci mapping to LTR retrotransposons are significantly down regulated following conditioning. In addition to the well characterized classes of short non coding RNAs, we describe a set of transcripts that produce large numbers of reads with a broad size distribution from the sense strand. We find that a subset of these are regulated following treatment and contain consensus elements that may be involved in their regulation. We investigate expression of one such gene with dramatically up-regulated reads following treatment, the Drosophila beta-site APP-cleaving enzyme (dBACE), and find that increased reads reflect increased mRNA levels. Further, we find that the target of dBACE protein, drosophila β amyloid protein precursor-like (APPL), is subjected to increased cleavage following conditioning, and that dBACE is required for LTM formation, but not for learning or STM. / Biology, Molecular and Cellular

Myelodysplastic syndromes (MDS) are a class of myeloid malignancy characterized by peripheral blood cytopenias and impaired hematopoietic differentiation. Our understanding of the molecular basis of MDS has improved enormously in recent years due to clinical research efforts to characterize the spectrum of acquired mutations found in patients. This work has revealed that mutations in TET2 are common lesions in MDS and other myeloid malignancies. TET2 function has only recently been elucidated: TET proteins convert 5’-methylcytosine (mC) first to 5’-hydroxymethylcytosine (hmC), apparently the first step in an active DNA demethylation program that leads to the replacement of 5-mC with unmodified cytosine. My thesis work focuses on a characterization of TET2 loss on DNA methylation, and on how TET2 mutations impact patient response to treatment with hypomethylating agents. We examined DNA methylation in a matched set of TET2-WT and -mutant MDS samples, and found that loss of TET2 results in global hypermethylation. This global increase is due to gains in intragenic methylation, specifically localized to intron-exon boundaries. We then used clonal TF1 cell lines with CRISPR/Cas9-engineered TET2 mutations to examine global DNA hydroxymethylation. Loss of TET2 results in a global loss of 5-hmC. By aligning our methylation data with hydroxymethylation data from TET2-WT cells, we were able to identify direct TET2 targets. Because changes in mC/hmC with loss of TET2 appeared to localize to intron-exon boundaries, we investigated the effect of aberrant methylation on mRNA splicing in our TF1 cell system. TET2 loss resulted in an overall increase in exon skipping, consistent with published data on the effect of methylation on splicing, and hypermethylated regions were enriched for alternate splicing events. These findings suggest that the alterations in hematopoietic differentiation seen in TET2-mutant models are due to shifts in the expression of different mRNA isoforms rather than wholesale changes in gene expression. Our data show that loss of TET2 function results in region-specific gains in DNA methylation, and that these alterations affect mRNA splicing by promoting exon skipping. Finally, we have found that presence of TET2 mutations are positively associated with response to HMA therapy. / Medical Sciences