Many synapses within the central nervous system are highly sensitive and responsive to ethanol. Although the regulation of postsynaptic receptors by alcohol is well studied, the mechanisms underlying the presynaptic effects of alcohol to alter neurotransmitter release remain relatively unexplored. This dissertation addresses whether alcohol-induced changes in transcriptional activity can promote synaptic vesicle fusion and therefore, neurotransmitter release. To identify a transcriptional pathway by which ethanol can regulate neurotransmitter release, we first investigated the effects of acute alcohol on the expression of genes encoding for synaptic vesicle fusion machinery proteins that form the soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs) complex. The proteins in this complex reside on the vesicle membrane (synaptotagmin 1 and synaptobrevin/vesicle-associated membrane protein, which is also known as VAMP) and the plasma membrane (syntaxin 1 and synaptosomal associated protein of 25 kDa, which is also known as SNAP-25), and their interactions within the SNARE complex trigger vesicle fusion and neurotransmitter release. We found that ethanol treatment of mouse cortical neurons increased the mRNA and protein expression levels of a subset of SNARE complex proteins, including synaptotagmin 1 (Syt1) and one of the isoforms of synaptobrevin, VAMP2, but not the other isoform, VAMP1. The gene induction of Syt1 and Vamp2 by alcohol occurs via activation of the transcription factor heat shock factor 1 (HSF1), while HSF1 transcriptional activity had no effect on Vamp1 mRNA levels. We then investigated whether ethanol altered neurotransmitter release in cortical neurons, using whole-cell voltage clamp electrophysiology. We found that alcohol increased gamma-aminobutyric acid (GABA) release via HSF1, but had no effect on glutamatergic synaptic vesicle fusion. Collectively, these data indicate that alcohol induction of HSF1 transcriptional activity triggers a specific coordinated adaptation in GABAergic presynaptic terminals that ultimately results in increased GABA release. This molecular mechanism could explain some of the transient changes in synaptic function that occur after alcohol exposure, and may underlie some of the enduring effects of chronic alcohol drinking on local circuitry.

Stroke has been a major source of morbidity and mortality for centuries. Eight-five percent of all strokes are ischemic in nature, meaning they are caused by the occlusion of a major cerebral artery. Despite extensive research to develop effective treatments for ischemic stroke, therapeutic options remain limited. Apoptosis (also termed "programmed cell death") is a process by which a stressed or damaged cell commits "suicide". In stroke, runaway apoptosis contributes to stroke neurodegeneration and neurological decline for days to weeks after disease onset. Cysteine-ASPartic proteASEs (caspases) are key mediators of apoptosis that are activated in distinct molecular pathways, but their impact in stroke is poorly defined. Direct evidence for caspase activation in stroke and the functional relevance of this activity has not been previously characterized. For this dissertation, we developed an unbiased technique for in vivo trapping of active caspases in rodent models of ischemic stroke. We isolated active caspase-9 as a principal contributor to ischemic neurodegeneration in rodents (Rattus norvegicus and Mus musculus). Caspase-9 is the initiator caspase for the intrinsic cell death pathway. Intranasal delivery of a novel, cell membrane-penetrating inhibitor for caspase-9 confirmed the pathogenic relevance of caspase-9 activity in stroke. Caspase-9 inhibition provided neurofunctional protection and established caspase-6 as its downstream target. Caspase-6 is an effector caspase and a member of the intrinsic death pathway that has never been implicated in stroke until now. Coincidentally, we discovered that caspase-6 is specifically activated within the axonal compartment. The temporal and spatial pattern of activation demonstrates that neuronal caspase-9 activity induces caspase-6 activation, which mediates axonal loss in the early stages of stroke (+/- 24 hours). We developed a novel inhibitor for caspase-6, based on a catalytically inactive clone, which demonstrated neuroprotective and axoprotective efficacy against ischemia. Collectively, these results assert that selective inhibition of caspase-9 and caspase-6 is an effective translational strategy for stroke. The impact of caspase activity is not restricted to neuronal death, as caspases can exacerbate inflammation and alter glial function. Thus, caspases are logical therapeutic targets for stroke. However, they have never been clinically evaluated due to a paucity of ideal drug candidates. This dissertation outlines fresh insights into the mechanisms of stroke neurodegeneration and offers novel caspase-based therapeutic strategies for clinical evaluation.

Stroke has been a major source of morbidity and mortality for centuries. Eight-five percent of all strokes are ischemic in nature, meaning they are caused by the occlusion of a major cerebral artery. Despite extensive research to develop effective treatments for ischemic stroke, therapeutic options remain limited. Apoptosis (also termed "programmed cell death") is a process by which a stressed or damaged cell commits "suicide". In stroke, runaway apoptosis contributes to stroke neurodegeneration and neurological decline for days to weeks after disease onset. Cysteine-ASPartic proteASEs (caspases) are key mediators of apoptosis that are activated in distinct molecular pathways, but their impact in stroke is poorly defined. Direct evidence for caspase activation in stroke and the functional relevance of this activity has not been previously characterized. For this dissertation, we developed an unbiased technique for in vivo trapping of active caspases in rodent models of ischemic stroke. We isolated active caspase-9 as a principal contributor to ischemic neurodegeneration in rodents (Rattus norvegicus & Mus musculus). Caspase-9 is the initiator caspase for the intrinsic cell death pathway. Intranasal delivery of a novel, cell membrane-penetrating inhibitor for caspase-9 confirmed the pathogenic relevance of caspase-9 activity in stroke. Caspase-9 inhibition provided neurofunctional protection and established caspase-6 as its downstream target. Caspase-6 is an effector caspase and a member of the intrinsic death pathway that has never been implicated in stroke until now. Coincidentally, we discovered that caspase-6 is specifically activated within the axonal compartment. The temporal and spatial pattern of activation demonstrates that neuronal caspase-9 activity induces caspase-6 activation, which mediates axonal loss in the early stages of stroke (<24 hours). We developed a novel inhibitor for caspase-6, based on a catalytically inactive clone, which demonstrated neuroprotective and axoprotective efficacy against ischemia. Collectively, these results assert that selective inhibition of caspase-9 and caspase-6 is an effective translational strategy for stroke. The impact of caspase activity is not restricted to neuronal death, as caspases can exacerbate inflammation and alter glial function. Thus, caspases are logical therapeutic targets for stroke. However, they have never been clinically evaluated due to a paucity of ideal drug candidates. This dissertation outlines fresh insights into the mechanisms of stroke neurodegeneration and offers novel caspase-based therapeutic strategies for clinical evaluation.

BioMEMS technology has the potential to increase the efficiency of conventional biological and medical protocols, by reducing their consumption of time and resources. Through more efficient surface-based chemical reactions and automation of tedious manual processes, orders of magnitude increases in efficiency across a number of metrics can be achieved by shifting conventional medical and biological protocols to the microscale domain. The SELEX process, by which aptamer sequences are selected via isolation from randomized libraries, is a time-consuming and resource-intensive protocol which is being performed with increasing frequency in both academic and private sector laboratories. Conventional approaches using macroscale technology cannot meet the current demand for selection of new aptamer sequences, as they require months of work and liters of expensive reagents. Microscale approaches to the SELEX process have been receiving attention in recent years due to their initial successes in reducing the time and reagents necessary to find aptamers. In particular, microscale "selection" or partitioning of weakly bound sequences from aptamer candidates, and on-chip integration of the protocol have separately been explored as approaches to scaling and improving SELEX. Initial results have shown that this technology can reduce resource requirements for SELEX by at least an order of magnitude. In this dissertation, a new approach to on-chip SELEX is developed which integrates highly efficient microfluidic selection and on-chip integration of the entire protocol. As a result, further reductions in processing time and reagent requirements can be realized. A demonstration of aptamer capabilities is first achieved via the development of a microfluidic aptasensor for cocaine, which utilizes aptamer-coated microbeads and fluorescent detection. Secondly, a technology necessary for on-chip integration of SELEX is developed: a novel bead-based polymerase chain reaction (PCR) protocol which vastly simplifies procedures for the capture and resuspension of ssDNA in solution. This protocol is then integrated on-chip with bead-based partitioning of weakly bound sequences to develop a microchip which performs temperature-specific isolation of aptamer sequences from a randomized library. Finally, this approach is further developed into a microfluidic SELEX chip which is capable of performing multiple rounds of temperature-specific SELEX. The novel bead-based protocol is shown to efficiently isolate target-binding sequences from a random library in a fraction of the time previously reported. As a result, this research provides a schematic for the development of highly efficient, integrated microfluidic SELEX devices. Such devices have the potential to impact a variety of fields including medical diagnostics, drug detection, and aptamer-based therapeutics.

The skeleton is a highly dynamic organ that undergoes constant remodeling to renew itself and maintain bone mass. It is subject to regulation from both hormones produced in peripheral tissues and neuronal control by the nervous system. Recent studies have shown that the skeleton is also an endocrine organ, releasing the hormone osteocalcin that increases insulin secretion and sensitivity in the pancreas and testosterone production in the Leydig cells of the testis. In my thesis study, I explore both the cell-nonautonomous regulation of osteoblast differentiation and the bone regulation of energy metabolism using cell-specific gene inactivation in the mouse. Early B-cell factor 1 (Ebf1) is a transcription factor whose inactivation in all cells results in high bone mass because of an increase in bone formation. To test if Ebf1 regulates bone formation cell-autonomously, I analyzed pattern of expression and its function in osteoblasts. I show here that in vivo deletion of in osteoblast progenitors does not affect osteoblast differentiation or bone formation accrual post-natally, indicating that the phenotype described in Ebf1 mice is not osteoblast-autonomous. Insulin signaling in osteoblasts contributes to whole body glucose homeostasis in the mouse and in humans by increasing the activity of osteocalcin. The osteoblast insulin signaling cascade is negatively regulated by ESP, a tyrosine phosphatase dephosphorylating the insulin receptor. is one of many tyrosine phosphatases expressed in osteoblasts, and this observation suggests that other protein tyrosine phosphatases may contribute to the attenuation of insulin receptor phosphorylation in this cell type. In this study, we sought to identify additional PTP(s) that like ESP, would function in the osteoblast to regulate insulin signaling and thus affect activity of the insulin-sensitizing hormone osteocalcin. For that purpose, we used as criteria, expression in osteoblasts, regulation by isoproterenol, and ability to trap the insulin receptor in a substrate-trapping assay. Here we show that the T-cell protein tyrosine phosphatase (TC-PTP) regulates insulin receptor phosphorylation in the osteoblast, thus compromising bone resorption and bioactivity of osteocalcin. Accordingly, osteoblast-specific deletion of TC-PTP (Ptpn2) promotes insulin sensitivity in an osteocalcin-dependent manner. This study increases the number of genes involved in the bone regulation of glucose homeostasis.

In eukaryotes, nuclear DNA is folded with histone proteins in the form of chromatin, and this structure plays a critical role in multiple biological processes, including development, DNA damage repair, and aging. Post-translational modifications of histones, such as acetylation and methylation, are essential regulators of chromatin structure and function. Consequently, misregulation of these post-translational modifications has causal roles in numerous diseases, including multiple types of cancer. However, the mechanisms that direct the localization of histone-modifying enzymes and regulate their activities are not fully understood. This thesis focuses on the characterization of a class of proteins containing the PWWP domain. This domain is often present in chromatin proteins, and it is predicted to recognize methylated histones based on structural analysis. Here, we have demonstrated that the PWWP domain proteins in fission yeast bind to methylated histones. Additionally, we have shown that proteins with this domain form complexes with diverse histone modifying activities to regulate multiple cellular processes. Methylation of histone H4 lysine 20 (H4K20me) is essential for the activation of a DNA damage checkpoint, which blocks the progression of cell cycle to allow sufficient time for DNA damage repair. In fission yeast, only the enzyme Set9 catalyzes H4K20me, and the mechanisms that underlie the regulation of this protein are poorly characterized. Here we showed that Set9 forms a stable complex with the PWWP domain containing protein Pdp1. The PWWP domain of Pdp1 binds to H4K20me, demonstrating that the PWWP domain constitutes a novel methyl-lysine recognition motif. Moreover, the binding of PWWP domain to methylated H4K20 plays a critical role in regulating Set9 activity, thus facilitating higher degrees of H4K20 methylation. Histone H3K9 methylation is critical for heterochromatin assembly in diverse organisms. The RNAi pathway is required for the formation of pericentric heterochromatin, although the exact role that RNAi plays in heterochromatin assembly remains a topic of significant debate. We discovered that a separate PWWP domain protein, Pdp3, forms a stable complex with the H3K14 histone acetyltransferase Mst2. Interestingly eliminating the enzymatic activity of the Pdp3-Mst2 complex obviates the requirement for the RNAi machinery in pericentric heterochromatin functions. Furthermore we demonstrated that one function of RNAi during heterochromatin assembly is to exclude the Pdp3-Mst2 complex, thus maintaining low levels of RNA polymerase II localization to pericentric regions in order to retain the parental histone modification patterns for its passage through generations. Altogether, my results have firmly demonstrated that the PWWP domain is a novel class of methyl-lysine binding motifs. Moreover, in fission yeast the PWWP domain proteins form stable complexes with other chromatin proteins to regulate diverse cellular processes.

Mutations within BRCA1 often contribute to breast cancer susceptibility. Many of these mutations cluster within two highly conserved regions that may be important for BRCA1's functions: the RING domain and the BRCT repeats. BRCA1's RING domain has E3 ubiquitin ligase activity, which is greatly enhanced when it forms a heterodimer with Bard1. This region is of particular interest because it displays the only known enzymatic activity of BRCA1. The BRCT repeats have phosphopeptide binding activity, which is necessary for BRCA1's interaction with DNA repair proteins BACH1, ABRAXAS, and CtIP. To test the importance of these two domains we generated cell lines and mouse models with point mutations that either eliminate the E3 ligase activity of Brca1 while maintaining its interaction with Bard1 (I26A) or eliminate its phosphopeptide binding activity (M1717R). We found that the E3 ubiquitin ligase activity of Brca1 is dispensable for its role in cell viability, embryonic development, double strand break repair, and tumor suppression. Interestingly, we were unsuccessful at generating homozygous BRCT mutant ES cells and homozygous M1717R Brca1 MEFs displayed a proliferation defect, spontaneous chromosomal aberrations, and centrosomal amplification. Our data shows that the BRCT repeats are crucial for BRCA1's role in DNA repair because BRCT mutant MEFs do not recruit Brca1 or Rad51 to IR induced DNA damage sites, they have a defect in homology directed repair, and M1717R Brca1 is not hyperphosphorylated in response to DNA damage in these cells. We were able to generate homozygous M1717R Brca1 mice, but with a very low frequency (<1%). All Brca1M17171R/M1717R mice produced thus far have been males and they are sterile. Our analysis indicates that the fertility defect is not due to a defect in meiosis. Introduction of the M1717R mutation in a conditional mammary or pancreatic tumor model also reduces the tumor latency to the same degree as the introduction of a Brca1 null mutation. Therefore our data shows that BRCA1 carries out the majority of its functions through its BRCT repeats.

During angiogenesis, nascent blood vessels sprout from pre-existing vasculature and recruit pericytes to induce maturation and vessel quiescence. Perictyes are associated with small vessels and capillaries where they share the basement membrane with the endothelium to provide vascular support. Pericytes are a critical component of the blood-brain barrier and regulate endothelial cell proliferation, vessel diameter, and vascular permeability. Endothelial cells express Notch1, whereas pericytes express both Notch1 and Notch3. Here we show that Notch signaling is essential for pericyte function. Through genetic manipulation and pharmacological tools we show that Notch regulates pericyte recruitment and pericyte/endothelial cell interactions. Notch1^+/-;Notch3^-/- mutant mice display decreased pericyte coverage and altered pericyte association with the retinal vascular plexus. Notch deficiency is associated with vascular anomalies where Notch1^+/-;Notch3-/- mice display retinal arteriovenous malformations (AVM) characterized by dilated vessels, vascular tangles and arteriovenous shunts that are similar to human brain AVMs. Disruption of pericyte/endothelial cell association is accompanied by an increase in vascular density, venule enlargement, and increased vascular permeability observed prior to AVM formation. In the ovary, we show that Jagged is essential for pericyte association with the endothelium where inhibition of Jagged-specific Notch activation results in luteal vessel dilation and hemorrhaging following ovarian hyperstimulation. By in vitro analysis of cultured pericytes we show that Notch1 and Notch3 induce plated derived growth factor receptor-β (PDGFR-β) expression to regulate cell migration. These findings expand the role for Notch in angiogenesis by demonstrating that Notch signaling in pericytes is essential for vascular development and function.

The dual specificity transcription factor Mga contains both a T-box binding domain and a basic helix-loop-helix zipper (bHLHZip) domain. Loss of Mga leads to embryonic lethality by E5.5. In vitro blastocyst culture and embryonic stem (ES) cell culture identify a lack of pluripotent inner cell mass (ICM) derived cells as the cause of embryonic lethality. Loss of Mga leads to increased apoptosis in E4.5 embryos, though there is no decrease in the amount of cell proliferation. Embryos with mutant Mga have fewer pluripotent ICM cells during delayed implantation, though the number of differentiated primitive endoderm cells remained initially stable. Despite the loss of pluripotent cells, there is no change in the pattern of expression of Nanog or Oct4, pluripotent cell markers, or Gata4, a primitive endoderm marker. Expression of Ornithine Decarboxylase (ODC), the rate-limiting enzyme in the synthesis of cellular polyamines, was identified as a possible cause of embryonic lethality based on a similar mutant phenotype as well as the presence of E-box sequences in genetic regulation loci. ODC is expressed at lower levels in the ICM of Mga mutants. Blastocyst and ES cell culture defects were rescued when cultured in the presence of exogenous putrescine, the metabolic product of ODC. These results suggest a mechanism for Mga to influence pluripotent cell survival through interactions with other bHLHZip domain proteins in the regulation of the polyamine pool in pluripotent cells of the embryo.

Binding proteins (both natural and man-made) have the ability to bind tightly and specifically with small molecules and other biopolymers. Binding proteins can function as human therapeutics, diagnostics, and as tools for scientific research. Given the wide range of potential applications, there is great interest in both academia and industry to develop methods for discovering novel binders. An important step in discovering new binders is called affinity maturation, when an initial hit that shows some ability to bind the target is further mutated in additional steps to improve binding affinity, specificity, solubility, pharmacokinetic profile. Ideally, the methods for affinity maturation would allow for cookbook protocols, be successful for arbitrary targets of interest, and be minimally resource intensive. Although traditional methods for affinity maturation have had some stunning successes over the past, the state of the field is still far from this ideal. In Chapter 1, I discuss the current state of the protein engineering field. In Chapter 2, I discuss the use of phenotypic selection for yeast-display protein binders, and test these systems on simple loop libraries. In Chapter 3, I construct a genetic system in yeast that can mutate a protein loop via homologous recombination, and test its recombination function. In Chapter 4, I mate libraries that target two different loops, and run FACS on the combinatorial libraries. In Chapter 5, I discuss future directions for the project.