Thesis (Ph. D.)--Ohio State University, 2006. / Title from first page of PDF file. Includes bibliographical references (p. 178-197).

To develop an assay to detect antibodies to Bacillus thuringiensis subsp. israelensis in mallard ducks, a growth curve was first established for the bacterium. The growth curve indicated that the crystal delta endotoxin would be best harvested from the rest of the cell material after 12 hours of growth. The delta endotoxin was solubilized in alkaline conditions followed by treatment with proteases or no treatment. The two differently treated delta endotoxins were purified by column chromatography. Fractions were assayed for duck erythrocyte lysis and cytotoxicity to a mosquito cell line. The proteolyzed sample gave four protein peaks with gel filtration, and the fourth peak containing biological activity was further separated into three protein fractions by anion exchange chromatography; two of the three showed biological activity. These two fractions contained 22 and 23 kD proteins species. The nonproteolyzed sample was separated into two protein fractions by gel filtration; only the first peak contained the biological activity. This fraction was further separated into two fractions by anion exchange chromatography; only the second fraction, containing a 28 kD protein, exhibited the activity. This fraction contained a 28 kD protein. However, the fractions containing 22 or 23 kD proteins originating from the proteolyzed sample showed the highest biological activity. Mallard ducks were repeatedly exposed to an aerosolized commercial preparation of the organism. Sera were collected periodically and tested for the antibody by an enzyme-linked immunosorbent assay (ELISA). Those toxic antigens containing 22 or 23 kD proteins were unsuitable for the assay. The exposed ducks were found to produce antibodies against the first fraction from anion exchange chromatography of the proteolyzed sample. The antibody titres increased as the number of exposures increased. The results suggest that ELISA is applicable for detecting antibodies against B.t.i. in wild ducks using the fraction containing a 50 kD protein. / Graduation date: 1992

Plant growth-promoting rhizobacteria (PGPR) are free-living, soil-borne bacteria that colonize the rhizosphere and, when applied to crops, enhance the growth of plants. Plant growth-promoting rhizobacteria may enhance plant growth either by direct or indirect mechanisms. The direct mechanisms of action include nitrogen fixation,production of phytohormones and lowering of ethylene concentrations. The objective of this study was to determine whether Pseudomonas putida strain 6-8 isolated from the rhizosphere of legume crops grown in Saskatchewan fields was able to promote the growth of canola cv. Smart and lentil cv. Milestone plants by direct mechanisms. Initial studies determined the effect of strain 6-8 and other known phytohormoneproducing PGPR strains on the growth of canola and lentil plants both in gnotobiotic and growth chamber conditions. Variations in the results were observed, as there were significant differences among trials. Strain 6-8 enhanced the growth of canola cv. Smart in growth pouches but not in pots in growth chamber studies. In the case of lentil cv.Milestone, strain 6-8 had no significant effect in growth pouches, but it significantly increased root dry weight, shoot dry weight and root surface area in pots in growth chamber studies. A similar effect was observed with wild-type strains GR12-2 and G20- 18. Strain GR12-2 was consistent in promoting the growth of lentil cv. Milestone both in growth pouches and in pots in growth chambers when compared to other strains and the control. The ability of the PGPR strains to produce auxin and cytokinin phytohomones in pure culture and in the canola rhizosphere was tested using the enzyme linked immunosorbent assay (ELISA). All the PGPR strains produced indole compounds and the concentration of the indoles produced increased with increasing concentrations of the precursor tryptophan. There were no significant differences among PGPR strains in production of indole-3-acetic acid (IAA) when assayed using ELISA. The concentrations of IAA secreted by PGPR strains were extremely low (0.19 µg/ml 9.80 µg/ml). Strain 6-8 produced the cytokinins, isopentenyl adenosine (IPA), zeatin riboside (ZR) and dihydroxyzeatin riboside (DHZR) in pure culture. Indole-3-acetic acid was detected in supernatants obtained from canola growth pouches inoculated with PGPR strains, but there were no significant differences in the concentrations of IAA secreted among PGPR strains. Significantly higher concentrations of IPA and ZR were observed in the rhizosphere of canola inoculated with strain 6-8 than in the non-inoculated control. Strain 6-8 produced siderophores, solubilized inorganic phosphate and used 1-aminocyclopropane-1-carboxylic acid (ACC), the precursor of ethylene, as sole nitrogen source. These traits are considered to be alternative mechanisms for direct plant growth promotion. A qualitative and quantitative study of root colonization by strain 6-8 was conducted by tagging the strain with green fluorescent protein in conjunction with confocal laser scanning microscopy and by conventional plating. The populations of strain 6-8 were higher on canola roots than on lentil roots by conventional plating. Similar results were also observed in confocal laser scanning microscopy (CLSM) studies after 5, 7 and 9 days for canola and 3, 6 and 9 days for lentil. Pseudomonas putida strain 6-8 produced cytokinins and also possessed other direct growth promoting characteristics. The ability of strain 6-8 to promote the growth of canola cv. Smart in growth pouches and lentil cv. Milestone in growth chamber studies may be related to these direct growth promoting characteristics. Strain 6-8 may have potential for development as a plant growth-promoting rhizobacterial inoculant.

<p> Bacterial cell division is an essential process, which is initiated by forming the Z-ring as a cytoskeletal scaffold at the midcell site, followed by the recruitment of a series of divisome proteins. In <i>Escherichia coli</i> (Ec), at least 15 divisome proteins (FtsZ, FtsA, ZipA, FtsK, FtsQ, FtsB, FtsL, FtsI, FtsW, FtsN, FtsE, FtsX, ZapA, AmiC, EnvC) have been implicated in this process. The components of the cell division machinery proteins in <i>Neisseria gonorrhoeae</i> (Ng) differs from <i>E. coli. N. gonorrhoeae</i> possesses FtsA, but lacks FtsB. ZipA and FtsL in <i>N. gonorrhoeae</i> have low identity to ZipA and FtsL from <i>E. coli</i>. Our laboratory has studied the central division protein FtsZ in <i>N. gonorrhoeae</i>. Thus, my research investigated the role of <i>N. gonorrhoeae</i> FtsA in cell division and investigated the interactions between divisome proteins from <i>N. gonorrhoeae</i> to understand divisome assembly.</p> <p>This study determined the association of FtsA_Ng with FtsZ</sub>Ng and other divisome proteins in <i>N. gonorrhoeae</i> and identified the functional domains of FtsA<sun>Ng</sub> involved in these interactions using a bacterial two-hybrid (B2H) assay. FtsA_Ng interacted with FtsZ_Ng, FtsK_Ng, FtsW_Ng, FtsQ_Ng, and FtsN_Ng. Self-interactions of FtsA_Ng and FtsZ_Ng were also detected. FtsI_Ng, FtsE_Ng and FtsX_Ng did not interact with FtsA_Ng. The 2A₁, 2A₂ and 2B domains of FtsA_Ng were sufficient to interact with FtsZ_Ng independently. Domain 2A₁ interacted with FtsK_Ng and FtsN_Ng. Domain 2B of FtsA_Ng interacted with FtsK_Ng, FtsQ_Ng, and FtsN_Ng. Domain 2A₂ of FtsA_Ng interacted with FtsQ_Ng, FtsW_Ng, and FtsN_Ng. These data suggest that FtsA in <i>N. gonorrhoeae</i> plays a key role in interactions with FtsZ and other divisome proteins.</p> <p>The potential interactions between divisome proteins in <i>N. gonorrhoeae</i> were examined using B2H assays. The comparisons between the <i>N. gonorrhoeae</i> divisome protein interaction network and those of <i>E. coli</i> and <i>S. pneumoniae</i> indicates that the divisome protein interactome of <i>N. gonorrhoeae</i> is more similar to that of <i>S. pneumoniae</i> and differs from that of <i>E. coli</i>. The comparisons revealed that compared to the interactions in <i>E. coli</i> and <i>S. pneumoniae</i>, more interactions between divisome proteins upstream of FtsA_Ng (including FtsA_Ng) and downstream of FtsA_Ng were observed in <i>N. gonorrhoeae</i> while fewer interactions between divisome proteins downstream of FtsA_Ng were observed in <i>N. gonorrhoeae</i>. Possible reasons for this include the inability of ZipA_Ng to interact with other divisome proteins and the absence of FtsL and FtsB in <i>N. gonorrhoeae</i>, resulting in the lack of an FtsQ-FtsB-FtsL complex in <i>N. gonorrhoeae</i>. These results indicate a possibly different divisome assembly in <i>N. gonorrhoeae</i> from that proposed models for <i>E. coli</i>.</p> A model for FtsA_Ng structure was predicted based on structural homology modeling with the resolved crystal structure of <i>Thermotoga maritima</i> FtsA. Four domains on the molecule were identified, designated 1A, 1C, 2B and 2A (including 2A₁ and 2A₂). Domains 2A and 2B of FtsA were highly conserved based on multi-sequence alignments of FtsAs from 30 bacteria. FtsA_Ng located to the division site in <i>N. gonorrhoeae</i> cells and the ratio of FtsA to FtsZ ranged from 1:24 to 1: 33 in three <i>N. gonorrhoeae</i> strains, which gave a lower cellular concentration of FtsA compared to other organisms.</p> <p>I also determined that overexpression of FtsA_Ng in <i>E. coli</i> led to cell filamentous in rod-shaped <i>E. coli</i> and cell enlargement and aggregation in mutant, round <i>E. coli</i>. FtsA_Ng failed to complement an <i>ftsA</i>_Ec-deletion <i>E. coli</i> strain although the overexperssion of FtsA_Ng disrupted <i>E. coli</i> cell division. In addition, overexpression of FtsA_Ng only affected cell division in some cells and its localization in <i>E. coli</i> was independent of interaction with <i>E. coli</i> FtsA or FtsZ. These results indicate that FtsA_Ng exhibits a species-specific functionality and <i>E. coli</i> is not a suitable model for studying FtsA_Ng functionality.</p> <p>This is the first study to characterize FtsA from <i>N. gonorrhoeae</i> in cell division. I identified novel functional domains of FtsA_Ng involved in interactions with other divisome proteins. The <i>N. gonorrhoeae</i> divisome protein interaction network determined by B2H assays provides insight into divisome assembly in <i>N. gonorrhoeae</i></p>.

Graduation date: 1992

The Saccharomyces cerevisiae Mms4 and Mus81 proteins are required for repairing DNA alkylation damage, but not damage caused by ionizing radiations. Previous studies have demonstrated that Mms4 and Mus81 form a specific complex in vivo, which functions as an endonuclease specific for branched DNA molecules. <p> In an effort to further understand the role of the Mus81-Mms4 complex in vivo, the structural and functional characteristics of this complex were investigated in this study. The epistatic analysis revealed that RAD52 was epistatic to MMS4 with respect to killing by methyl methanesulfonate (MMS), suggesting that MMS4 is involved in the RAD52 dependent homologous recombinational repair pathway. However, the mms4 rad51, mms4 rad54 and mms4 rad50 double mutants showed more sensitivity to MMS than either corresponding single gene disruptant. Since Rad51 and Rad54 are required to form the Holliday junction during recombinational repair pathway, it is unlikely that the Mus81-Mms4 complex functions as a Holliday junction resolvase in vivo. <p> The role of MMS4 in DNA damage induced mutagenesis has been investigated. Deletion of MMS4 had no obvious effects on damage-induced basepair mutations, but increased frame-shift mutations by 3 fold when the yeast cells were treated with MMS. This suggests that the Mus81-Mms4 complex plays a role in limiting the damage-induced frame-shift mutagenesis. <p> Through a yeast two-hybrid assay, Mus81 and Mms4 have been demonstrated to form a stable and specific complex in vivo. This result is consistent with previous studies. To localize the domains of the Mms4 and Mus81 proteins involved in herterodimer formation, a series of deletion mutants were constructed for the yeast two-hybrid assay. The Mus81-binding domain of Mms4 was mapped to the extreme C-terminal region between amino acids 598-691. The Mms4-binding domain of Mus81 was mapped to a domain between amino acids 527-632. The results from co-immunoprecipitation experiment were consistent with those from the yeast two-hybrid assay. The Mms4-1 (Gly173Arg) protein was found to lose its interaction with Mus81, and this kind of amino acid substitution is very likely to alter the three-dimension structure of the protein. Thus we hypothesize that the three-dimensional structure is also important for Mms4 to interact with Mus81. <p> By studies on green fluorescent protein (GFP) fusion proteins and their subcellular localization, we demonstrated that Mms4 and Mus81 are nuclear proteins. When the putative nuclear localization sequence 1 (residues 244-263) in Mms4 was deleted, the truncated protein lost the ability to enter the nucleus. On the contrary, deletion of the putative nuclear localization sequence 2 (residues 539-555) had no effect on the localization of the protein. Furthermore, the nuclear localization of Mus81 was proven to be independent of its interaction with Mms4, and the N-terminal half of Mus81 is necessary and sufficient for its localization to the nucleus.

The Saccharomyces cerevisiae Mms4 and Mus81 proteins are required for repairing DNA alkylation damage, but not damage caused by ionizing radiations. Previous studies have demonstrated that Mms4 and Mus81 form a specific complex in vivo, which functions as an endonuclease specific for branched DNA molecules. <p> In an effort to further understand the role of the Mus81-Mms4 complex in vivo, the structural and functional characteristics of this complex were investigated in this study. The epistatic analysis revealed that RAD52 was epistatic to MMS4 with respect to killing by methyl methanesulfonate (MMS), suggesting that MMS4 is involved in the RAD52 dependent homologous recombinational repair pathway. However, the mms4 rad51, mms4 rad54 and mms4 rad50 double mutants showed more sensitivity to MMS than either corresponding single gene disruptant. Since Rad51 and Rad54 are required to form the Holliday junction during recombinational repair pathway, it is unlikely that the Mus81-Mms4 complex functions as a Holliday junction resolvase in vivo. <p> The role of MMS4 in DNA damage induced mutagenesis has been investigated. Deletion of MMS4 had no obvious effects on damage-induced basepair mutations, but increased frame-shift mutations by 3 fold when the yeast cells were treated with MMS. This suggests that the Mus81-Mms4 complex plays a role in limiting the damage-induced frame-shift mutagenesis. <p> Through a yeast two-hybrid assay, Mus81 and Mms4 have been demonstrated to form a stable and specific complex in vivo. This result is consistent with previous studies. To localize the domains of the Mms4 and Mus81 proteins involved in herterodimer formation, a series of deletion mutants were constructed for the yeast two-hybrid assay. The Mus81-binding domain of Mms4 was mapped to the extreme C-terminal region between amino acids 598-691. The Mms4-binding domain of Mus81 was mapped to a domain between amino acids 527-632. The results from co-immunoprecipitation experiment were consistent with those from the yeast two-hybrid assay. The Mms4-1 (Gly173Arg) protein was found to lose its interaction with Mus81, and this kind of amino acid substitution is very likely to alter the three-dimension structure of the protein. Thus we hypothesize that the three-dimensional structure is also important for Mms4 to interact with Mus81. <p> By studies on green fluorescent protein (GFP) fusion proteins and their subcellular localization, we demonstrated that Mms4 and Mus81 are nuclear proteins. When the putative nuclear localization sequence 1 (residues 244-263) in Mms4 was deleted, the truncated protein lost the ability to enter the nucleus. On the contrary, deletion of the putative nuclear localization sequence 2 (residues 539-555) had no effect on the localization of the protein. Furthermore, the nuclear localization of Mus81 was proven to be independent of its interaction with Mms4, and the N-terminal half of Mus81 is necessary and sufficient for its localization to the nucleus.

This thesis presents consolidated research results of a low-cost, high efficiency, high throughput detection system for fluorescence-based bio-assays. Such high throughput screening process is an invaluable tool for the multifaceted field of Systems Biology, where it is widely used in genomics and proteomics for drug and gene discovery applications. The thesis is divided into three parts: addressing the feasibility of using hydrogenated amorphous silicon photodiodes as the sensor, the development of an associated compact model suitable for circuit-level simulations, and integration of the sensors and switches to realize the array. Requirements of fluorescent bio-assays demand low sensor dark current densities in the order of 10¯¹¹A/cm² at room temperature. Fabrication of high quality segmented a–Si:H n–i–p photodiodes with such specification is achieved by tailoring defects at photodiode junction sidewalls, where both the dry etching and passivation conditions play important roles. Measurements of the fabricated photodiodes at different temperatures allowed the extraction of reverse current components, which are necessary in modeling such sensors in Verilog-A. Two prototype array designs are fabricated with pixel dimensions matching ANSI standard microwell plates. The functionalities of the small arrays are demonstrated with green LEDs to simulate fluorescent dyes that are commonly used in the high throughput bio-assay processes.

Many of today’s methods for the detection of biomolecules suffer from a high limit ofdetection due to poor signal generation upon recognition of target. By applying andoptimizing proximity ligation assay (PLA) in Western blotting (WB), the limit of detectionhas been lowered down to the picomolar range. In this report I have optimized the differentparameters that affect the signal generation and explored possibilities to increase the ease ofuse, by merging protocol steps and performing signal generating reactions at roomtemperature.

Background: As a part of a national health control program, Statens VeterinärmedicinskaAnstalt performs diagnostics to screen flocks for certain pathogens causing high mortality,morbidity and/or serious economical losses. There are several viruses in the programincluding IBDV (infectious bursal disease virus), IBV (infectious bronchitis virus) and NDV(Newcastle disease virus). Method: 96 serum samples were collected from different poultryflocks in Sweden and analyzed by ELISA, which are currently used in the health controlprogram as well as by a commercial prototype of a multiplex immunoassay manufactured byLuminex Corp., which is currently under evaluation at the United States Department ofAgriculture USDA. This 4-plex assay detects antibodies for the three above-mentionedviruses as well as antibodies of avian reovirus. In the context of this study the ELISAs run inroutine diagnostics as well as a REO ELISA were used as the standard for comparison.Result: The antibody concentration in serum from vaccinated chickens was high while theantibody concentration level in serum from not vaccinated chickens was low. There was agood correlation between the multiplex assay and ELISA. Conclusion: The results obtainedin this study indicate that the Luminex multiplex assay tested has the potential to be usedroutinely for screening flocks.