The bacterium Saccharopolyspora erythraea is utilized to produce the widely used antibiotic erythromycin. With the increasing problem of antibiotic resistance, the development of derivatives of the drug is of great importance. To produce antibiotics biosynthetically, it is crucial to understand the metabolic pathways utilized by the bacterium. It is key to clarify the behaviour of the proteins that sense the environment and regulate the gene expression accordingly. The iron-dependent regulator IdeR utilizes iron as a cofactor and is the main transcriptional regulator acting on genes crucial for iron homeostasis. Its main function is to act as a repressor for genes that are necessary for iron uptake, but it also functions as an activator for genes that promote iron storage. The role of the protein’s SH3-like domain in gene regulation is not understood. However, it might serve as a crosslinker between dimers bound to adjacent binding sites and be necessary for proper gene regulation. The aim of this paper was to examine this hypothesis. Cysteine residues with high probability of forming disulfide bonds between the SH3-like domain and the rest of the protein were introduced to IdeR and the constructed variants characterized. Methods used included electrophoretic mobility shift assay (EMSA), circular dichroism (CD) spectroscopy, and small-angle X-ray scattering (SAXS). Attempts to characterize the proteins using differential scanning fluorimetry (DSF) and X-ray crystallography were also performed. Four IdeR variants were produced and purified. The EMSA analysis indicated that theiraffinities for the consensus DNA sequence was similar to or a bit lower than that of the wild-type (wt). The CD spectroscopy measurements showed that the variants had, similar to the wt, mainly α-helical secondary structure. One of the variants had, however, less secondary structure content than the wt and was probably less stable. The secondary structure content of the wt did not change upon metal binding. The variants’ melting temperatures under oxidizing conditions were similar to or lower than that of the wt. The same experiment in reducing environment showed a loss of cooperativity for all examined variants. No data was gained from the DSF experiment and the IdeR variants did not crystallize under the investigated conditions during the project timeline. The results from the SAXS experiment show that the IdeR variants had the same overall structure as the wt as dimers and when bound to DNA. Conclusions that can be drawn from the experiments are that the introduced mutations did not affect the overall structure nor the thermal stability of the protein. However, since it cannot be concluded from the performed experiments if the crosslink between the protein domains formed as intended, another attempt to get crystal structures of the proteins is required.

DNA double strand breaks represent the single most dangerous type of damage that can afflict the genome. Given the severity of such a lesion, higher eukaryotes possess two distinct pathways to repair such damages. The work presented here focuses on the role of different protein complexes formed during within Non-Homologous End Joining (NHEJ). Specifically, how the C-terminal tails of XRCC4 and XLF regulate higher-order complex formation of end bridging filaments prior to terminal ligation and release of the intact DNA following repair. The crystal structure of full length XRCC4 was solved to 3.43Å and confirmed that the C-terminal tails or XRCC4 mediate tetramerization but are not required for end bridging of DNA ends. A cluster of residues that stabilized the XRCC4 multi helix bundle were mutated and determined to result in an XRCC4Mutational analysis and SEC-MALS further revealed that this 4-helix bundle stabilizes tetramers, interestingly tetramerization was found to not be required for bridging of DNA ends. Additional work aimed at determining the mechanism by which XLF binds DNA and how complex filaments are formed was carried out using a combination of structural and biochemical techniques. Mutational analysis of the yeast XLF homologue, Nej1, revealed that the tails of these proteins bind their DNA substrates through an extended interface that may involve wrapping DNA to further stabilize interaction. Also, it was determined that phosphorylation of key residues within this extended DNA binding domain results in a decreased affinity for DNA and may play a role in DNA repair pathway choice in Saccharomyces cerevisiae. Transmission electron microscopy showed that when bound to DNA, XLF is capable of forming DNA dependent filaments that are capable of bridging DNA ends in a linear manner. Addition of XRCC4 resulted in extensive remodelling of these filaments. Crystals of the XRCC4/XLF/DNA were optimized to diffract to a resolution of greater than 5Å, however further work will be required to determine the structure of this key NHEJ complex. Finally, attempts to determine the optimal combination of DNA substrates and NHEJ factors to crystallize the terminal repairosome were carried out and initial hit conditions have been identified. / Thesis / Doctor of Philosophy (PhD)

Arsenic is a ubiquitous environmental toxic substance. As a consequence of continual exposure to arsenic, nearly every organism, from Escherichia coli to humans have evolved arsenic detoxification pathways. One of the pathways is extrusion of arsenic from inside the cells, thereby conferring resistance. The R773 arsRDABC operon in E. coli encodes an ArsAB efflux pump that confers resistance to arsenite. ArsA is the catalytic subunit of the pump, while ArsB forms the oxyanion conducting pathway. ArsD is an arsenite metallochaperone that binds arsenite and transfers it to ArsA. The interaction of ArsA and ArsD allows for resistance to As(III) at environmental concentrations. The interaction between ArsA ATPase and ArsD metallochaperone was examined. A quadruple mutant in the arsD gene encoding a K2A/K37A/K62A/K104A ArsD is unable to interact with ArsA. An error-prone mutagenesis approach was used to generate random mutations in the arsA gene that restored interaction with the quadruple arsD mutant in yeast two-hybrid assays. Three such mutants encoding Q56R, F120I and D137V ArsA were able to restore interaction with the quadruple ArsD mutant. Structural models generated by in silico docking suggest that an electrostatic interface favors reversible interaction between ArsA and ArsD. Mutations in ArsA that propagate changes in hydrogen bonding and salt bridges to the ArsA-ArsD interface also affect their interactions. The second objective was to examine the mechanism of arsenite resistance through methylation and subsequent volatilization. Microbial ArsM (As(III) S-adenosylmethyltransferase) catalyzes the formation of trimethylarsine as the volatile end product. The net result is loss of arsenic from cells. The gene for CrArsM from the eukaryotic green alga Chlamydomonas reinhardtii was chemically synthesized and expressed in E. coli. The purified protein catalyzed the methylation of arsenite into methyl-, dimethyl- and trimethyl products. Synthetic purified CrArsM was crystallized in an unliganded form. Biochemical and biophysical studies conducted on CrArsM sheds new light on the pathways of biomethylation. While in microbes ArsM detoxifies arsenic, the human homolog, hAS3MT, converts inorganic arsenic into more toxic and carcinogenic forms. An understanding of the enzymatic mechanism of ArsM will be critical in deciphering its parallel roles in arsenic detoxification and carcinogenesis.

<div> <div> <div> <p>Phages, viruses that infect bacteria, have been used for many studies in understanding fundamentals of molecular biology and taking advantage of their natural antimicrobial properties (Harper 2021). They are often noted for their overwhelming abundance and are recognized as the most abundant biological entities in the world (Harper 2021). The field has grown since the early 20th century, and now, there are several classes of phages that have been observed and characterized (Ackermann 2009). Within this abundant class of biological organisms, the order called Caudovirales, is the most populated group of phages to date (Harper 2021). In this order of viruses, the dsDNA genome phages have 2 main components, the icosahedral capsid, and a tail (Harper 2021). Though many tailed phages have been studied for many decades, new information about phages is still being found. Important findings such as the CRISPR gene editing tool adapted from phages in 2007 (Barrangou, Fremaux et al. 2007) have contributed to new biotechnology that impacts human health. For this reason, studies on phages have proven to be valuable in understanding fundamental biological questions and advancing basic research. </p><p><br></p> <p>In this dissertation, we investigated phage G, which has the largest capsid and genome of propagated phage studied to date (Donelli 1968, Sun and Serwer 1997, Pope 2011, Hua, Huet et al. 2017). By studying phage G, we may add to the knowledge of this relatively unexplored group of Jumbo phages with remarkably larger genomes (>200kbp) (Yuan and Gao 2017)to understand how their structure and function may be similar or different to the commonly studied, smaller bacteriophages, such as T4 and λ. For a majority of these studies, we outline how our structural biology insights of phage G using cryo-EM (cryo-Electron Microscopy) have shown it’s icosahedral capsid of ~ 180 nm in diameter at the 5-fold icosahedral vertex is composed of hexamer and pentamer proteins similar to what’s been discovered in other, smaller tailed phages (González, Monroe et al. 2020). Our observations from microscopy data also show unique mechanistic properties in phage G’s tail that are inconsistent with current model of tail contraction within the Myoviridae family of tailed phages. Data suggest phage G’s structure and organization of its helical tail are still similar to contractile phages such as T4 (Amos and Klug 1975, Abuladze, Gingery et al. 1994) and phi812 (Nováček, Šiborová et al. 2016), however, the mechanism of the tail sheath movement is inconsistent with the existing ideas of myophage function (Harper 2021).</p></div></div></div>

<p>The purpose of this longitudinal study was to determine whether a non-monotonic pattern characterizes the development of postural control in children from age 5 to 8 years, as suggested by previous cross-sectional studies. Postural control was considered in terms of control strategy and its variability operationalized by mean and standard deviation of Centre of Pressure velocity (COPvel); and of effectiveness and its variability operationalized by mean and standard deviation of COP anteroposterior excursion (YS). Periods of significant variability were used to indicate behavioural transitions. Seventeen, healthy children (9 males, 8 females) aged 5 to 6 years (61.5 - 75 months) were tested at 3 to 4 month intervals until age 8 years (83 - 97 months) in eyes-open quiet stance (OS) on a force platform for 30 seconds in each of 10 trials. Data were reorganized into 6 developmental categories based on adjacent test dates prior to (-2, -1) and after (+1, +2, +3) a subject's trial with the lowest COPvel. Developmental category is proposed to represent level of sensorimotor integrative skill. A 1-way MANCOVA revealed a significant effect ( p <.0001) for developmental category with covariance due to height, weight and actual age removed. Post-hoc 1-way ANCOVAs showed a significant effect (p<.OOO1) on measures of strategy. However differences in COPvel (type of strategy used) and differences in its variability (denoting a transition between types of strategies) were not always coincident. Performance outcome (yS) changed linearly across categories. From a consideration of the results it was concluded that a non-monotonic change in control strategy describes the development of quiet stance equilibrium. A transition, marked by variability of COPvel occurs from a primarily open-loop to incorporation of open- and close-loop components of control. Honing of strategy used precedes and follows transitions. Constriction of velocity and excursion may typify the early stages of bi-modal strategy use suggesting instability in realworld situations at this stage. Linear change in effective excursion regardless of strategy employed in unchallenged as may decrease the utility of this measure in assessing stability status in children. Developmental categories describe affiliation with the strategy employed and may represent differentiable levels of sensorimotor integrative skill. As such, they may be more useful in assessing progression of equilibrium control than consecutive age in years.</p> / Master of Science (MSc)

Master of Science (MSc)

<p>Pleckstrin is the major substrate of PKC in activated platelets, therefore a role for pleckstrin in mediating platelet responses such as secretion has been widely hypothesized. Previous cloning of the pleckstrin gene indicated that it encoded a unique protein sequence containing no obvious structural, catalytic or otherwise functional motifs, apart from several candidate PKC phosphorylation sites and homology between the Nand C terminal domains.</p> <p>The role of pleckstrin in mediating cytoskeletal changes that occur during platelet activation and secretion was investigated based on published data implicating pleckstrin in directly inhibiting actin polymerization in its unphosphorylated fonn. There were no significant effects on the kinetics of actin polymerization in vitro using recombinant pleckstrio or pleckstrin purified from platelets. In accordance with this observation, overexpression of pleckstrin in COS cells did not appear to affect the subcellular distribution or ftlament length of actin compared to uninfected cells. Also, pleckstrin did not significantly co-localize with actin prior to or after exposure to phorbol 12-myristate 13- acetate (PMA) concentrations known to induce maximal phosphorylation of pleckstrin. It was therefore concluded that pleckstrio does not playa direct role in regulating actin polymerization.</p> <p>Based on limited sequence homology to dynamin (which is involved in endocytosis), I postulated that pleckstrio might regulate vesicle motility, a key requirement for platelet degranulation. Pleckstrin immunofluorescently labelled in transfected COS cells did not appreciably co-localize with either ER or Golgi structures, the precursors of secretory vesicles, or with microtubules, however some punctate fluorescence was observed that co-localized with the ER. Immunfluorescent detection of pleckstrin in peripheral blood cells indicated that pleckstrin is most highly abundant in platelets, moderately expressed in polymorphonuclear cells and absent or lowly expressed in lymphocytes.</p> <p>Pleckstrin's prevailing characteristic, its phosphorylation, was examined in greater detail. Wild-type recombinant pleckstrin overexpressed in COS cells displayed a 5 fold stimulation of ³²P incorporation in response to PMA treatment. To exclude the possiblity of phosphorylation by kinases other than PKC, the effect of staurosporine on inhibiting pleckslrio phosphorylation was examined. Staurosporine inhibited pleckstrin phosphorylation in the heterologous COS cell system with an IC₅₀ = 10 nM. similar to that reported to inhibit pleckstrin phosphorylation in the platelet. Additionally. identical phosphopeptide maps were generated of wild-type pleckslrio isolated from HL-60, HEL or transfected COS cells indicating that pleckstrin expressed recombinantly in COS cells was by all known measures functionally competent. However, a pleckstrin mutant deleted for 42 residues including the most probable phosphorylation sites was de-regulated for phosphorylation, displaying a high basal level of phosphorylation, no response to PMA, and increased association with cytoskeletal or membraneous components.</p> <p>To precisely map the phosphorylation sites of pleckstrin. 8 serine or threonine residues were converted to alanine using site-directed mutagenesis. Two of these mutants, S113A and SI17 A, consistently displayed a lower level of PMA-induced hosphorylation compared to the wild-type protein. Tryptic phosphopeptide maps of three pleckstrin mutants, S1l3A, T1l4A and S1l7A indicated that phosphorylation of pleckstrin occurs almost exclusively on SI13 and S117; these sites accounted for 46 and 52% respectively of the incorporated phosphate. A minor, variable amount of phosphorylation on TI14 was observed. These phosphorylation sites were confirmed by phosphopeptide mapping synthetic peptides corresponding to residues 108-120 of pleckstrio, and by performing phosphoamino acid analysis on the tryptic phosphopeptides generated in these maps.</p> <p>Lastly, overexpression of each of the phosphorylation site mutants of pleckstrin in COS cells indicated that one of them, S 113A, might possess a dominant activity that perturbs membrane/cytoskeletal structures. This mutant induced large, semi-circular ruffles or "lobes" in transfected COS cells in two independent experiments.</p> <p>In conclusion, the function of pleckstrin is still entirely unknown. It is quite clear however that pleckstrin is an excellent substrate for PKC. Phosphorylation of pleckstrin likely affects its function, a hypothesis supported in this work by the unusual phenotype of COS cells overexpressing the Sll3A mutant.</p> / Doctor of Philosophy (PhD)

<p>The nature of the peroxisomal targeting signal in Candida tropicalis trifunctional enzyme, hydratase-dehydrogenase-epimerase (HDE) was investigated. The first part of this thesis describes the cloning and sequencing of the cDNA and gene encoding HDE. The second and third parts of the thesis describe the amino acid sequence (and general nature) of the signal that directs HDE to peroxisomes in yeast. The cDNA and gene encoding C. tropicalis HDE was isolated from a λgt11 cDNA expression library and from a λEMBL3 genomic library, respectively. Primary sequence analysis of both the cDNA and the gene revealed a single open reading frame of 2718 nucleotides encoding a protein of 906 amino acids (calculated molecular mass = 99,481 Da). The gene encoding HDE was expressed in the yeasts Candida albicans and Saccharomyces cerevisiae. The cellular location of HDE was determined by subcellular fractionation followed by western blot analysis of peroxisomal and cytosolic fractions using antiserum specific for HDE. HDE was found to be exclusively targeted to and imported into peroxisomes in both heterologous expression systems. Deletion and mutational analyses were used to determine the regions within HDE which are essential for its targeting to peroxisomes. Deletion of a carboxy-terminal tripeptide Ala-Lys-Ile completely abolished targeting of HDE to peroxisomes, whereas large internal deletions of HDE (amino acids 38-353 or 395-731) had no effect on HDE targeting to peroxisomes in either yeast. Substitutions within the carboxy-terminal tripeptide (Ala → Gly and Lys → Gln) supported targeting of HDE to peroxisomes of C. albicans but not of S. cerevisiae. Antiserum directed against the carboxy-terminal 11 amino acids of HDE, containing the Ala-Lys-Ile tripeptide (anti-AKI) was used to probe subcellular fractions from several yeasts and rat liver. The anti-AKI serum reacted exclusively with multiple peroxisomal proteins from the yeasts C. tropicalis, C. albicans and Yarrowia lipolytica suggesting a common motif for the targeting of some proteins to yeast peroxisomes, the prototype of which is the HDE carboxy-terminal tripeptide Ala-Lys-Ile.</p> / Doctor of Philosophy (PhD)

<p>Studies concerned with peroximes biogenesis have established that peroxisomal proteins are encoded exclusively by nuclear genes, synthesized (with few exceptions) on free polyribosomes at their mature size and posttranslationally imported into preexisting oganelles. From this, it was realized that the information responsible for the specificity of protein targeting to peroxisomes must reside within the mature amino acid sequence of the imported protein. Knowledge of the targeting sequences of peroxisomal proteins is, however, limited. Consequently, using the yeast C. tropicalis pK233 as a model system, this work endeavoured to clone and sequence a cDNA or gene encoding a peroxisomal protein so that the subsequent creation of deletion mutants of this cDNA/gene, together with the development of and in vitro and/or in vivo expression and translocation system for the corresponding polypeptides, would allow the amino acid sequence(s) comprising the peroxisomal addressing signal of this protein to be identified.</p> <p>Related efforts resulted in the cloning of two genes (and one composite full-length cDNA) encoding different enzymes associated with the C. tropicalis peroxisomes. Sequencing of the gene encoding acyl-coenzymes A oxidase (AOx), revealed a single open reading frame, without intervening sequences, of 2,127 nucleotides which encodes a polypeptide of 709 amino acids (M_r 79155). The deduced amino acid sequence of C. tropicalis AOx exhibits significant identity and homology with that of several other acyl-coenzyme A oxidases of C. tropicalis, as well as other yeasts (C. maltosa and S, cerevisiae) and rat. By comparison, sequencing of the C. tropicalis gene (or cDNA) encoding catalase (Cat) revealed a single open reading frame of 1,455 nucleotides, without intervening sequences, which encodes a polpeptide of 485 amino acid (M_r 54944). The deduced amino acid equence of C. tropicalis Cat shows sigificant identity and homology with that of catalases from the yeast S. cerevisiae mammals (rat, human and bovine), plants (maize, sweet potato and cottonseed) and Drosophila. COdon usage in the genes encoding C.tropicalis AOx and Cat is nonrandom, exhibiting a strong bias towards a group of preferred codons of S. cerevisiae and E. coli, as weel as other C. tropicalis genes encoding peroxisomal proteins.</p> <p>Having achieved the charaterization of the genes encoding C. tropocalis AOx and Cat, attempts were made to establish an in vitro system for the translocation of Cat into peroxisomes purified from C. tropicalis which could be used to determine the peroxisomal targeting signal (PTS) of Cat. For this purpose, N-terminal, C-terminal, N-/C-terminal and internal deletion mutants of Cat were constructed, expressed in cell-free systems and assayed for translocation into purified C. tropicalis perxisomes. Unfortunately, all attempts at in vitro imports proved to be inadequate for the translocation of Cat into purified C. tropicalis peroxisomes. Since a functional in vitro system could not be established, studies concerned with the in vivo expression and targeting of Cat from C. tropicalis to the peroxisomes of C. albicans and S. cerevisiae were initiated. Although, the in vivo expression of Cat sequences could not be attained in C. albicans, the expression of full-length and truncated Cat polypeptides was achieved in S. cerevisiae. Subsequent determination of the subcellular location of the full-length Cat polypeptide suggested that Cat from C. tropicalis was targeted to the peroxisomes of S. cerevisiae, reaffirming that heterologous signals for import into peroxisomes are recognized in S. cerevisiae. Interestingly, the deletion of amino acids 182 to 344 of Cat negated the targeting of Cat to S. cerevisiae peroxisomes. These preliminary in vivo studies suggest that at least part of the PTS of Cat from C. tropicalis resides at an internal location within its mature amino acid sequence.</p> / Doctor of Philosophy (PhD)

Phleboviruses are emerging zoonotic pathogens which constitute a global threat to human and animal health. The mosquito-borne Rift Valley fever virus (RVFV) is a widespread problem across the African continent and causes regular deadly outbreaks in ruminants. The recently emerged severe fever with thrombocytopenia syndrome virus (SFTSV) is a serious human public health concern in China which has rapidly spread to Japan and Korea with fatality rates as high as 16-30%. Phleboviral cell entry is mediated by two viral glycoproteins: the class II fusion protein Gc and the lesser known Gn. Initial cell attachment is glycan dependent and the penetration into the cell cytoplasm is mediated by the Gc fusion protein which catalyses viral and cell membrane merger. The entry mechanism is not well understood from a structural perspective which limits mechanistic insights. The purpose of this thesis is to further our understanding of the cell entry process by filling in the missing structural information on the phleboviral glycoprotein layer. To this end, an integrated structural approach using cryo-EM and X-ray crystallography was adopted. The crystal structure of the Gn ectodomain is presented which reveals an unprecedented structural relationship with seemingly unrelated viruses. Single-particle cryo-EM and localized reconstruction reveal the glycoprotein layer of the RVFV and a pseudo-atomic model of the RVFV is presented. The assembly shows the shielding of the Gc fusion protein and suggests that the Gn functions as a fusion chaperone. The post-fusion crystal structure of the Gc protein from SFTSV further consolidates a mechanism of membrane fusion by class II fusion proteins. Finally, preliminary data on receptor binding and mechanism of antibody mediated neutralization are presented. The work presented herein provides a novel platform for studying and understanding entry and assembly of phleboviruses as well as the design of novel therapeutics.