Electrostatics interactions are fundamental for protein stability and function, and play a crucial role upon protein-DNA binding. Both the protein and the DNA from a complex exhibit clear shape and electrostatic complementarity. Counterions from the solvent surround the DNA in order to neutralize its highly anionic charges. The interplay between the delocalized binding and release of counterions has been originally established by the counterion condensation theory (CCT). The correlation between the binding constant Kobs and the salt concentration [M+] provides the measurement SKobs. CCT also states that the number of ionic contacts formed in the interface of protein and DNA during binding equals SKobs. In this work we explore the validity of the above relationship and we question the efficacy of SKobs descriptions for larger, more complex protein-DNA systems. We also analyzed the role of the charge distribution in protein stability and potentially in binding affinity between protein and DNA. We employ a computational approach that accounts for the non-specific salt-mediated electrostatic interactions by measuring the electrostatic free energy of binding with the Poisson-Boltzmann (PB) equation. We investigated the distinct salt-dependent binding behavior of halophilic TATA-binding proteins (TBP) to DNA and of four families of DNA-binding proteins (homeo-domains, high mobility groups, interferon regulatory factors, and basic-region leucine zippers). We were able to obtain a PB-based definition of SK for the protein-DNA complexes (SKpred), which compares to SKobs, except in cases when protein and/or DNA undergo dramatic conformational changes. We did not observe the previously determined correlation between SK and ion pairs devised by CCT for any of the protein-DNA complexes in this study, and therefore we believe this correlation does not hold for complexes larger than the original oligocationpolyelctrolytes tested. Since the correlation was determined for a series of oligocationpolyelectrolyte complexes, where the DNA was represented by a short line of charges (instead of the double stranded conformation of DNA), the correlation fails to capture the molecule's structural information. Besides, the correlation between SK and ion pairs only accounts for protein cations as ion pair forming. For that reason, we decided to test the effect of anions over SKpred. By neutralizing the charge of anionic residues at the binding interface of TBP and of the other proteins of the four DNA-binding protein families, we noticed a severely altered SKpred. In some cases, we performed surface mutations on TBP, in order to reverse the charge of anionic residues at TBP's binding interface, in which case the halophilic nature of the protein approaches a mesophilic-like behavior when bound to DNA. Combined, these results indicate the crucial effect of charge distribution into SKpred and potentially to the the stability of a protein-DNA complex. Our studies led to the development a protein redesign program that can automatically generate protein mutants, in an effort to achieve protein structures with higher thermal stability. We created a de novo protein design algorithm based on a Monte Carlo method coupled with a replica exchange approach. We validated the usefulness of our approach in repacking a large group of diverse proteins and also compared the effect of different parameters (such as force field and rotamer libraries) upon the final resulting structures. We observed that our approach is system-independent, and still eficient for very large proteins. Our algorithm was employed in the modeling of more stable variants of fibroblast growth factor-1 protein. Based on the work presented here, we believe that current interpretations of SKobs will have to be reconsidered, and we are currently working to expand CCT. We observed that the effects from the charge distribution are much more complex and can relate to protein stability, as seen in the thermophilic TBP. The charged mutations obtained from our protein design algorithm could help to engineer proteins with higher thermostability. Newly redesigned proteins could potentially be more resistant to degradation or have longer shelf-life; therefore aiding the fields of industry and medicine. / A Dissertation Submitted to the Institute of Molecular Biophysics in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Summer Semester, 2010. / May 3, 2010. / DNA-Binding Proteins, Electrostatics, Salt-Dependence, Protein Design / Includes bibliographical references. / Marcia O. Fenley, Professor Co-Directing Dissertation; Michael Blaber, Professor Co-Directing Dissertation; Michael Mascagni, University Representative; Hugh Nymeyer, Committee Member; Timothy Logan, Committee Member.

Cell adhesion assemblies occur at sites where cells either contact each other or components related to the extracellular matrix. They provide the structural integrity needed to support nonmigrating cells via a host of transmembraneous proteins. In addition to their structural role, these transmembrane receptors also establish a system of communicating with the cytoskeleton. One of the primary receptors found on the surface of stationary cells is the fibronectin receptor, also known as the a5b1-integrin. When bound to their extracellular ligand, fibronectin, these integrins assume an " active" conformation. This external binding event is coordinated with a series of physical changes that are translated through the transmembrane portion of the receptor and passed along to its cytoplasmic domain. Inside the cell, these perturbations cause a series of structural changes that allow the F-actin cytoskeleton to be linked to the integrin's cytoplasmic domain. Our interests lie in visualizing the macromolecular assemblies of the cytoskeletal components that support this linkage. The method we used to observe these adhesive complexes is transmission electron microscopy (TEM). We used the lipid monolayer crystallization technique for a dual purpose: 1) as a means of concentrating protein at the air:water interface, providing a crystallization surface; 2) to mimic the cytoplasmic leaflet, which contains a hydrophobic lipid layer on top of a bulk aqueous phase. Therefore, preparing an ordered EM specimen with these characteristics gives us a tool to study numerous biological systems. In this project specifically, we synthesized the integrin cytoplasmic domain with a histidine (His)-tag at its N-terminus and bound it to a lipid monolayer containing a nickel-chelating group. This causes the integrin to assume an orientation that represents its native conformation at the cell membrane. We were able to produce EM specimens that contained this integrin domain along with other cytoskeletal proteins, such as talin, a-actinin, vinculin and F-actin. In particular, the b1-integrin:a-actinin and the b1-integrin:aactinin: vinculin samples formed ordered arrays that we used to make frozen-hydrated specimens. We collected EM images of untilted samples and calculated their averaged 2-D projections. Additionally, we produced a 3-D reconstruction of the b1-integrin:a-actinin:vinculin assembly by collecting images of samples tilted up to 70o. Using molecular modeling software, in combination with information deposited into various protein databases, we created atomic models of both the b1-integrin as well as the vinculin-head piece. The docking of these models into our 3-D Cryo-EM map was quantified and refined to produce an atomic model for our assembly. Overall, this combination of imaging and model building establishes a methodology of producing complexes of adhesive proteins whose spatial relationships are virtually unknown. / A Dissertation submitted to the Institute of Molecular Biophysics in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester, 2003. / March 27, 2003. / Intracellular and Extracellular Scaffolding, Adhesion / Includes bibliographical references. / Kenneth A. Taylor, Professor Directing Dissertation; Thomas M. Roberts, Outside Committee Member; Michael Blaber, Committee Member; Kenneth H. Roux, Committee Member; Thomas C. S. Keller, III, Committee Member.

The differences in the concepts of two different parameterizations of cloud microphysics are analyzed. Simulations alternatively applying these parameterizations are performed for a Baltic heat cyclone event. The results of the simulations are compared to each other as well as to observed distributions of cloudiness and snowfall. The main differences between the simulated distributions result from the assumptions on ice, the ice classes, and size distributions of the cloud and precipitating particles. Both schemes succeeded in predicting the position and the main structure of the main cloud and snowfall fields. Nevertheless, the more convective type parameterization overestimates, while the other one underestimates snowfall. / Die Unterschiede in den Konzepten zweier unterschiedlicher Parametrisierungen der Wolkenmikrophysik werden analysiert. Die Ergebnisse der Simulationen werden miteinander und mit den beobachteten Wolken- und Schneeverteilungen für eine Baltische Wärmezyklone verglichen. Die wesentlichen Unterschiede in den berechneten Verteilungen resultieren aus den verschiedenen Annahmen über Wolkeneis, die Eisklassen und die Größenverteilungen der Wolken- und Niederschlagspartikel. Beide Schemata sagen die Position und die wesentlichen Strukturen der Wolken- und Schneeverteilungen erfolgreich vorher. Dennoch überschätzt das eher konvektive Schema den Schneefall, während das andere ihn unterschätzt.

This dissertation explores the electronic characteristics of silicon at the single molecule level. This idea is born as we enter the post-Moore’s law era when the exponential shrinking of conventional silicon microelectronics has begun to stall and an investigation of molecular materials is timely. Single-molecule electronic components have shown promising functionalities such as conductors, switches, and diodes, and single molecule junctions have become a widely used test-bed for probing electron transport properties at the molecular level. In this thesis, we use scanning tunneling microscope break junction method to create single molecule junctions with a variety of silicon molecular wires. Our results demonstrate electronic properties of silicon beyond it being a semiconductor in its bulk form. We begin this work in pursuit of an expanded understanding of low-k dielectric components with an experimental goal on determining the cause of its breakdown. Low-k dielectrics are beneficial as they enable faster switching speeds and lower heat dissipation, however, they tend to breakdown after prolonged usage under an applied voltage. At the atomic level, low-k dielectric breakdown involves bond rupture. To determine which bond breaks easily, we conduct experimental studies on the robustness of individual chemical bonds that are commonly found in low-k dielectrics. We subject the single molecule junctions to a high bias and investigate the breakdown phenomenon of individual Si-Si, Ge-Ge, Si-O, and Si-C bonds. Among these, Si-C proved to be significantly more durable than the others. To further prove our hypothesis that the Si-Si bond ruptures under the applied high bias, we design a two-path molecular structure consisting of a Si-Si bond in parallel with a naphthyl group. The broken junction shows conduction through the naphthyl pathway, strongly indicating that the Si-Si bond is breaking. This demonstrates a method for probing the bond cleavage under an electric field and provides insights to the weak links in low-k dielectrics. Next, we study the fundamental charge transport characteristics of single molecule junctions comprised of Si and Ge-based molecular wires, starting with the simplest form - linear atomic chains. We observe a slower decay of conductance with increasing length in the silanes and germanes than in alkanes, indicating that the electronic delocalization in the Si-Si and Ge-Ge -bonds is stronger than that of the well-studied C-C bonds. Furthermore, we demonstrate that this electronic delocalization in the Si-Si and Ge-Ge bonded backbones enables single-molecule conductance switching. This conductance switch, induced by a mechanical modulation, relies on the nature of the terminal groups and constitutes the first example of a stereoelectronic switch. We also study the molecular conductance of these silanes with metal contacts other than Au, which can potentially open up interesting avenues as metal varies in its electronic states and catalytic activities. We find that Ag electrodes enable higher conductance for thiol-terminated silanes than Au or Pt electrodes. The electrical properties of more complex silicon structures - silicon rings - were probed. We choose a five-membered silicon ring as a target system to investigate the effect of isomerism on single molecule conductance. We find that due to the flexibility of the ring, multiple conformations contribute to the spread in the measured conductance for each isomer. This provides us with a starting point to further compare the conductance of a variety of silicon rings. We find that most of the silicon rings are less conductive than their linear counterparts due to the suboptimal backbone conformation for electronic coupling. In particular, destructive quantum interference appears in one of the bicyclic structures and leads to an exceptionally low conductance. This is the first example of a destructive quantum interference feature ever observed experimentally in a π-bonded rather than a σ-bonded system. Finally, we investigate the impact of strain on molecular conductance of silanes. In one case, we introduce the strain using a silacyclobutane ring in the backbone. Unexpectedly, we find that ring strain enables a new Au-silacycle binding mode, resulting in a much higher conductance state. In another molecular design, we choose disilaacenaphthene in the backbone. This strained disilane is found to constitute an example of a direct Si-to-Au contact in single molecule circuits, thereby demonstrating a new binding motif that is valuable for designing high conducting molecular components. Taken together, this body of work provides important knowledge about the transport properties of silicon at the nano-scale, as well as insights on the design of silicon components for nanoelectronics. This work represents one step forward to create functional silicon molecular components.

This thesis forms a NERC funded CASE studentship with the Met Office, whose aim is to investigate the treatment of cloud microphysical processes in numerical models, with a particular focus on exploring the impacts and possible benefits of microphysical complexity for the purpose of simulating clouds and precipitation. The issue of complexity is an important one in numerical modelling in order to maintain computational efficiency, particularly in the case of operational models. The latest numerical modelling tools are utilised to perform simulations of cloud types including idealised trade wind cumulus, orographic wave cloud and wintertime shallow convective cloud. Where appropriate, the modelling results are also validated against observations from recent field campaigns. The Factorial Method is employed as the main analysis tool to quantify the effect of microphysical variables in terms of their impact on a chosen metric. Ultimately it is expected that the techniques and results from this thesis will be used to help inform the future development of cloud microphysics schemes for use in both cloud resolving and operational models. This is timely given the current plans to upgrade the microphysics options available for use within the Met Office Unified Model. For an idealised warm cloud, it is shown that different bin microphysics schemes can produce different results, and therefore additional microphysical complexity does not necessarily ensure a more consistent simulation. An intercomparison of bin microphysics schemes in a 1-D column framework is recommended to isolate the origin of the discrepancies. In relation to the mixed-phase wave cloud, model simulations based on an adaptive treatment of ice density and habit struggled to reproduce the observed ice crystal growth rates, highlighting the need for further laboratory work to improve the parameterization of ice growth by diffusion within the sampled temperature regime. The simulations were also found to be largely insensitive to values of the deposition coefficient within the range of 0.1 to 1.0. Results from a mesoscale modelling study of shallow wintertime convection demonstrate the importance of the representation of dynamical factors that control cloud macrostructure, and how this has the potential to overshadow any concerns of microphysical complexity. Collectively, the results of this thesis place emphasis on the need to encourage more synergy between the dynamics and microphysics research communities in order to improve the future performance of numerical models, and to help optimise the balance between model complexity and computational efficiency.

We use a lipid monolayer system to prepare paracrystalline rafts of F-actin with two different cross-linking proteins, α-actinin and villin. α-Actinin cross-links F-actin into relatively loose networks throughout cells, while villin forms dense, tight bundles of filaments in specialized structures known as microvilli. Our monolayer system allows us to examine each of these cross-linkers as they interact with F-actin in a simplified two-component system. In order to analyze the data from these arrays, we have hybridized methods from 2-D crystal analysis with a single-particle approach to refine strategies for correspondence analysis and classification of images. For the α-actinin:F-actin rafts, we use this strategy to characterize the highly variable cross-links as 2-D projection images. The villin:F-actin rafts require further refinement of the hybrid strategy by application to 3-D volumes from electron tomography. Both protein arrays yield unique insights to the architectural arrangement of cross-linking proteins between filaments. In the α-actinin:F-actin rafts we use correspondence analysis to demonstrate that otherwise polar arrays of F-actin can have insertions of filaments of reversed polarity within them, and that these polarity differences do not influence the cross-links. We developed a method for left-right independent classification of the α-actinin cross-links to recover the high variability in the cross-link angular distribution by increasing the signal-to-noise ratio of the class averages. These averages are combined to recreate the original cross-link as it appears in a process we call "mapping-back." Measurements reveal that the length of cross-links can vary. From the classification of cross-links we demonstrate and model the novel occurrence of α-actinin bound to successive cross-over repeats on the same actin filament which we have termed "monofilament" binding. We further illustrate that the length variation of these monofilament-bound α-actinins are quantized to 55 Å, the distance between two adjacent actin monomers in the filament. The villin:F-actin rafts displayed homogeneous cross-linking. Docking of the homologous gelsolin atomic coordinates into the three-dimensional volumes reveals that villin does not interact with F-actin in the same manner as gelsolin. This data supports multiplicity of binding modes even in highly homologous proteins, and suggests a new mode of F-actin-binding for the villin protein. / A Dissertation submitted to the Institute of Molecular Biophysics in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester, 2006. / August 11, 2006. / Cytoskeleton, Electron Tomography, Actin, Alpha-Actinin, Villin, Correspondence Analysis, Lipid Monolayer, Electron Microscopy / Includes bibliographical references. / Kenneth A. Taylor, Professor Directing Dissertation; Charles Ouimet, Outside Committee Member; P. Bryant Chase, Committee Member; Piotr Fajer, Committee Member; Hong Li, Committee Member.

Tuberculosis, an infectious disease caused by Mycobacterium tuberculosis, has been a leading cause of death in the world infecting over one third of the world population (WHO 1999). A detailed structural understanding of the membrane proteins and cell envelope of M. tuberculosis will help unravel the mechanism through which bacteria infect and survive within human macrophages, and thus may help to facilitate the development of new treatments for tuberculosis. In fact, it is anticipated that more than 50% of all drug targets are membrane proteins. However, there is very little structural information on these proteins. Our goal for this project is to characterize the three-dimensional backbone structure of Rv0008c, an integral membrane protein from M. tuberculosis by solution Nuclear Magnetic Resonance (NMR) spectroscopy. Rv0008c is a small, conserved membrane protein containing 145 residues and one transmembrane helix. It is in a gene cluster that we have shown to form a complex with Rv0011c - also known as a homologue of CrgA, an FtsZ inhibitor - a critical component of the cell division machinery. A detailed three-dimensional structure of this protein will help to understand its critical role in cell division process and therefore this will help to suggest a possible cell division control mechanism in M. tuberculosis. Controlling cell division is important since M. tuberculosis goes into a "latent" state where no cell division occurs and none of the current drugs are effective against this state. If M. tuberculosis can be prevented from entering the "latent" state, the drug regimen can be shortened dramatically. Backbone assignments for Rv0008c were made using 2D and 3D- NMR experiments on uniformly 13C/15N–labeled Rv0008c protein in DPC (dodecyl phosphocholine) micelles. Dihedral angle restraints were derived from chemical shifts using TALOS (Torsion Angle Likelihood Obtained from Shifts and sequence similarity) program. Two sets of residual dipolar couplings were measured for proteins weakly aligned in compressed neutral and positive charged gels. Long-range paramagnetic relaxation enhancement (PRE) distances were obtained for protein samples at six spin label sites. NOE (nuclear Overhauser effect) distance restraints were also obtained to improve the quality of the structure. The three-dimensional backbone structure of Rv0008c was calculated and refined by the Xplor-NIH program using the dihedral angle, residual dipolar couplings, PRE and NOE distance restraints. / A Dissertation submitted to the Program in Molecular Biophysics in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester, 2008. / April 9, 2008. / Membrane Protein, Mycobacterium Tuberculosis, Backbone Structure, Functional Genomics, Structural Genomics, Cell Division, Solution Nmr / Includes bibliographical references. / Timothy A. Cross, Professor Directing Dissertation; P. Bryant Chase, Committee Member; Timothy M. Logan, Committee Member; Samuel Grant, Committee Member; Penny Gilmer, Outside Committee Member.

An ever-growing body of evidence suggests that dark matter exists and is abundant in our universe. Although the direct detection of dark matter has yet to be realized, the intensity of the experimental and theoretical search continues to amplify. The question is no longer whether dark matter exists, but rather what is its fundamental nature and how can it be known. Many large-scale, international experiments are actively searching for one class of dark matter candidates, weakly interacting massive particles (WIMPs). While indirect searches, such as those looking for the creation of dark matter in particle accelerators or for the Standard Model byproducts of dark matter annihilation, are contributing significantly to our understanding of the properties WIMPs may have, direct searches, such as those using cryogenic liquids and solids to look for scattering, have produced the most stringent limits on the properties of WIMPs. Liquid xenon (LXe) detectors continue to lead the field in the search for the direct detection of WIMPs. The success of experiments using LXe relies upon decades of measurements of the fundamental properties of LXe itself, as well as thorough characterization of the detectors that utilize this amazing element. One frontier of LXe detectors is at low energies. Next-generation LXe detectors, such as XENON1T, require a better understanding of the response of LXe to particle interactions as a function of electric field, as well as more precise measurements of the radioactive backgrounds that contribute to low-energy electronic recoil interactions. In this thesis, I describe details of efforts to characterize the stability of the XENON100 detector during its primary dark matter search periods in 2010-2012. I examine the electronic recoil data for any indications of periodic behavior, and compare the XENON100 result with a dark matter annual modulation claim by DAMA/LIBRA. I also describe the design, construction, and performance of a dedicated experiment to measure the low-energy properties of LXe, in particular the energy and electric field dependence of the response of LXe to electronic recoils. Finally, I describe the design and performance of an atom trap trace analysis device for assaying the levels of radioactive krypton in LXe dark matter detectors.

Clouds play an important role in precipitation, solar radiation budget and electrification of Earth's atmosphere. The presence of small ice crystals in clouds and their morphology can complicate parametrisation and climate modelling, consequently leading to a net cooling feedback on climate. In situ airborne measurements provide single particle characterisation with high temporal and spatial resolution allowing better understanding of atmosphericprocesses of ice nucleation and growth. Simulations of the preindustrial clouds and accurate characterisation and comparison of the instruments require a well-controlled and often pristine environment. The experimental chamber setup allows simulations of these and other conditions. The microphysical features of the micrometric ice particles in clouds were examined in a laboratory setup, at numerous sub-zero temperatures [-10 to -50 ⁰C]. The following instruments were sampling the content of the CLOUD chamber air volume: Cloud and Aerosol Spectrometer with Polarisation (CASPOL), Particle Phase Discriminator mark 2 (PPD-2K, Karlsruhe edition), 3-View Cloud Particle Imager (3V-CPI), and the Scattering-Intensity-Measurements-for-the-Optical-detectioN-of-icE (SIMONE-Junior). Cluster analysis was applied to the data collected with CASPOL and compared with results from the other probes. We were able to discriminate and map the aerosol and cloud particles in the pristine chamber environment using polarisation ratios (Dpol/Backscatter and Dpol/Forwardscatter) of the scattered light. We demonstrate the sensitivity of the instruments in detecting secondary organic aerosol (SOA) phase transitions. Then, we show the ability of the viscous SOA to nucleate ice in a series of SPectrometer for Ice Nuclei (SPIN) measurements. The detected viscous SOA ice nucleation efficiency may affect global modelling and estimations of ice water content in the atmosphere. Subsequently, the analysis and discrimination technique used in the CLOUD chamber was applied to airborne measurements to test its efficiency and to retrieve the composition of clouds. Data from four flight campaigns on board of the FAAM BAe-146 were analysed: Aerosol-Cloud Coupling and Climate Interactions in the Arctic (ACCACIA), COnvective-Precipitation-Experiment (COPE) in south England, CIRrus Coupled Cloud-Radiation EXperiment (CIRCCREX), and PIKNMIX in Scotland. In these and other flights, we were able to identify unique clusters such as salts, minerals, organics, volcanic ash, water and ice, confirming some of the offline laboratory elemental analysis results, and providing complementary information. Single particle polarisation measurements were compared with bulk depolarisation, diffraction patterns, and imaging. Most of the optical instruments still suffer from ambiguity in phase derivation (i.e. water/ice) of optically spherical small shapes. We discuss some of the limitations of optical cloud particle discrimination in different ambient conditions and offer possible solutions to reduce the uncertainty, e.g., surface complexity derivation from scatteringpatterns. Our findings will help to develop better instruments and improve the models which are used for weather forecasts and climate change predictions.

Modern protein catalysts are often viewed as possessing exquisite specificities for their cognate physiological substrates. In contrast, primordial catalysts are thought to have possessed much broader substrate specificities, a characteristic that likely afforded the survival of their host organisms under a plethora of diverse environmental conditions. Recent experimental work suggests that present day enzymes often retain the ability to recognize and transform a variety of natural and unnatural compounds that are structurally distinct from their target substrate. The widespread existence of such promiscuity could prove generally useful both in the natural and directed evolution of new proteins. To probe the persistence of enzyme promiscuity in modern proteomes we studied the model organism Escherichia coli due to its rapid growth, ease of genetic manipulation and many years of prior research on this organism which have generated abundant knowledge on its metabolism. The first exploration into uncovering enzyme promiscuity, described in chapter two, examines the proton transfer reaction catalyzed by triosephosphate isomerase (TIM). Triosephosphate isomerase catalyzes the interconversion of D-glyceraldehyde 3-phosphate and dihydroxyacetone phosphate, an essential step in glycolytic and gluconeogenic metabolism. To uncover promiscuous isomerases embedded within the E. coli genome, we searched for genes capable of restoring growth of a TIM-deficient bacterium under gluconeogenic conditions. Rather than discovering an isomerase, we selected yghZ, a gene encoding for a member of the aldo-keto reductase superfamily. Here we show that YghZ catalyzes the stereospecific, NADPH-dependent reduction of L-glyceraldehyde 3-phosphate, the enantiomer of the TIM substrate. This transformation provides an alternate pathway to the formation of dihydroxyacetone phosphate. In chapter three we show that Gpr co-purifies with a b-type heme cofactor. Gpr associates with heme in a 1:1 stoichiometry to form a complex that is characterized by a Kd value of 5.8 ± 0.2 µM in the absence of NADPH and a Kd value of 11 ± 1.3 µM in the presence of saturating NADPH. The absorbance spectrum of reconstituted Gpr indicates that heme is bound in a hexacoordinate low-spin state under both oxidizing and reducing conditions. The physiological function of heme association with Gpr is unclear, as the L-glyceraldehyde 3-phosphate reductase activity of Gpr does not require the presence of the cofactor. Bioinformatics analysis reveals that Gpr clusters with a family of putative monooxygenases in several organisms, suggesting that Gpr may act as a heme-dependent monooxygenase. The discovery that Gpr associates with heme is interesting because Gpr shares 35% amino acid identity with the mammalian voltage-gated K+ channel β-subunit, an NADPH-dependent oxidoreductase that endows certain voltage-gated K+ channels with hemoprotein-like, O2-sensing properties. To date the molecular origin of O2 sensing by voltage-gated K+ channels is unknown and the results presented herein suggest a role for heme in this process. In chapter four we probe the network of genes within E. coli that can provide resistance to the nonnatural toxin bromoacetate. Microbial niches contain toxic chemicals that are capable of forcing organisms into periods of intense natural selection to afford survival. Elucidating the mechanisms by which microbes evade environmental threats has direct relevance for understanding and combating the rise of antibiotic resistance. In this study we used a toxic small-molecule, bromoacetate, to model the selective pressures imposed by antibiotics and anthropogenic toxins. We report the results of genetic selection experiments that identify nine genes from Escherichia coli whose overexpression affords survival following exposure to a lethal concentration of bromoacetate. Eight of these genes encode putative transporters or transmembrane proteins, while one encodes the essential peptidoglycan biosynthetic enzyme, UDP-N-acetylglucosamine enolpyruvoyl transferase (MurA). Biochemical studies demonstrate that the primary physiological target of bromoacetate is MurA, which becomes irreversibly inactivated via alkylation of a critical active-site cysteine. Genetic experiments also identify 63 single-gene mutants of E. coli that display increased susceptibility to bromoacetate. One hypersensitive bacterium lacks yliJ, a gene that encodes a glutathione transferase capable of catalyzing the detoxification of bromoacetate with a kcat/Km value of 5.4 × 103 M-1 s-1. The catalytic proficiency of YliJ, which exceeds 5 orders of magnitude, is particularly noteworthy considering the enzyme is unlikely to have previously encountered bromoacetate. In total, our results indicate that nearly 2% of the E. coli proteome contributes to, or can be recruited to provide, bromoacetate resistance. This illustrates the wealth of intrinsic survival mechanisms that can be exploited by bacteria when they are challenged with toxins. The work described here illuminates the vast metabolic and functional plasticity of protein function harbored within bacteria. Their ability to recruit latent and weakly active proteins for novel functions enables survival under diverse nutritional and environmental challenges. / A Dissertation submitted to the Department of Chemistry and Biochemistry in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester, 2010. / June 14, 2010. / Enzyme Function, Antibiotic Resistance, Glutathione Transferase / Includes bibliographical references. / Brian Miller, Professor Directing Dissertation; Hank Bass, University Representative; Hong Li, Committee Member; Lei Zhu, Committee Member; M. Elizabeth Stroupe, Committee Member.