abstract: Extremal graph theory results often provide minimum degree conditions which guarantee a copy of one graph exists within another. A perfect $F$-tiling of a graph $G$ is a collection $\mathcal{F}$ of subgraphs of $G$ such that every element of $\mathcal{F}$ is isomorphic to $F$ and such that every vertex in $G$ is in exactly one element of $\mathcal{F}$. Let $C^{3}_{t}$ denote the loose cycle on $t = 2s$ vertices, the $3$-uniform hypergraph obtained by replacing the edges $e = \{u, v\}$ of a graph cycle $C$ on $s$ vertices with edge triples $\{u, x_e, v\}$, where $x_e$ is uniquely assigned to $e$. This dissertation proves for even $t \geq 6$, that any sufficiently large $3$-uniform hypergraph $H$ on $n \in t \mathbb{Z}$ vertices with minimum $1$-degree $\delta^1(H) \geq {n - 1 \choose 2} - {\Bsize \choose 2} + c(t,n) + 1$, where $c(t,n) \in \{0, 1, 3\}$, contains a perfect $C^{3}_{t}$-tiling. The result is tight, generalizing previous results on $C^3_4$ by Han and Zhao. For an edge colored graph $G$, let the minimum color degree $\delta^c(G)$ be the minimum number of distinctly colored edges incident to a vertex. Call $G$ rainbow if every edge has a unique color. For $\ell \geq 5$, this dissertation proves that any sufficiently large edge colored graph $G$ on $n$ vertices with $\delta^c(G) \geq \frac{n + 1}{2}$ contains a rainbow cycle on $\ell$ vertices. The result is tight for odd $\ell$ and extends previous results for $\ell = 3$. In addition, for even $\ell \geq 4$, this dissertation proves that any sufficiently large edge colored graph $G$ on $n$ vertices with $\delta^c(G) \geq \frac{n + c(\ell)}{3}$, where $c(\ell) \in \{5, 7\}$, contains a rainbow cycle on $\ell$ vertices. The result is tight when $6 \nmid \ell$. As a related result, this dissertation proves for all $\ell \geq 4$, that any sufficiently large oriented graph $D$ on $n$ vertices with $\delta^+(D) \geq \frac{n + 1}{3}$ contains a directed cycle on $\ell$ vertices. This partially generalizes a result by Kelly, K\"uhn, and Osthus that uses minimum semidegree rather than minimum out degree. / Dissertation/Thesis / Doctoral Dissertation Mathematics 2019

No description available.

Graduation date: 2002

Previous reports indicate the presence of multiple CYP1A sequences in rainbow trout, but their functional differences are unknown. This report describes the cloning and partial characterization of four trout CYP1A cDNAs, which are given the tentative designations CYP1A1v2, v3, v4, and v5. Comparison among these four and three previously reported trout CYP1A sequences reveals that all of the nucleotide and translated amino acid sequences all are closely related (96.9-99.4% cDNA identity; 95.2-99.4% amino acid sequence identity) but none are identical. Six of these sequences encode proteins of 522 amino acids, and one encodes a protein of 536 amino acids. Expression vectors containing the cDNAs for CYP1A1v2, v3, and v4 were transformed into yeast, yielding microsomal hemoprotein CYP contents (63, 156, 96 pmol/mg) comparable to those reported for human CYP1A1 (68-156 pmol/mg) expressed in this system (Eugster et al., 1990, Biochem. Biophys. Res. Commun. 737-744). Kinetic analysis of CYP1A1v2 and v3 proteins indicated similar but not identical Michaelis constants (20��3 vs 13��2 ��M) and molar activities (508��47 vs 218��19 pmol/min/nmol P450) for oxidation of aflatoxin B��� (AFB���) to aflatoxin M a reaction characteristic of human CYP1A2. Trout CYP1A1v2 and v3 exhibited lower activity for production of AFB,-8,9-exo-epoxide, also a human CYP1A2 activity. Kinetic data for ethoxyresorufin 0deethylation, a prototypical mammalian CYP1A1 activity, also revealed modest but distinct differences in which CYP1A1v3 was more active for this substrate (Km=0.07 �� 0.01 ��M, Vm=1398 �� 95 pmol/min/nmol P450) than was CYP1A1v2 (Km=0.15 �� 0.03 ��M, Vm=684 �� 83 pmol/min/nmol p450). Interestingly, CYP1A1v4 showed no catalytic activity towards AFB ethoxyresorufin, or 7,12-dimethylbenzanthracene despite formation of a hemoprotein. These results together with previous studies demonstrate the presence in various rainbow trout populations of at least seven CYPIA cDNAs representing gene duplication or allelic variation. Present results show that one of three such cDNA sequences encodes a CYP1A hemoprotein with no apparent catalytic activity, that two of the encoded proteins possess certain catalytic properties common to both human CYP1A1 and CYP1A2, and that the sequence differences, though small, are reflected in enzymic properties that can be distinguished. / Graduation date: 1998

Currently little is known about the mechanisms and locations of lymphocyte development in teleosts. In this study several aspects of the underlying factors which govern B lymphocyte development in trout were investigated which included: the isolation and characterization of immunoglobulin heavy chain (IgH) genes, the recombination activating genes 1 and 2 (RAG1 and RAG2) and the use of cellular markers to identify tissues harboring precursor B-cells. Immunoglobulin heavy chains are part of the structural components which make up antibody molecules produced by B-cells. We isolated various full-length IgH cDNA clones, some of which contained the secreted while others contained the membrane bound form of IgH. Upon characterization of the membrane bound forms, typical features common to all IgH cDNAs were found including a leader peptide, a variable region and constant domain containing transmembrane (TM) segments as well. Further sequence analysis of this region revealed that the TM domains were spliced directly to the CH3 domains which results in the loss of the entire CH4 region. Our results support previous observations of unusual splicing events in fish IgH genes. RAG1 and -2 in mammals have been shown to be essential for carrying out V (D) J recombination of lymphocyte receptors and are found to be expressed within primary lymphoid tissues and precursor lymphocytes. We isolated the RAG locus from a rainbow trout genomic library and characterized their conservation and expression. Overall the complete amino acid sequences of RAG1 and RAG2 displayed 78% and 75% similarity when compared to RAG genes from higher vertebrates thus demonstrating the highly conserved nature of these genes. Tissue specific expression of both genes was primarily associated with the thymus and pronephros in both juvenile and adult trout. Based upon these observation we conclude that the thymus and pronephros likely serve as the tissue sites for V (D) J recombination in trout and are thus primary lymphoid organs. Finally we addressed the question as to where B-cell lymphopoiesis occurs in trout. Our results using both immunofluorescence and confocal microscopy putatively demonstrate that the thymus harbors precursor B-cells and thus alludes to a dual function for both B and T-cell development in trout. / Graduation date: 1996

Graduation date: 1996

Graduation date: 1995

Graduation date: 1995

The organochlorine pesticide endosulfan has been shown to be highly toxic to fish and there is some evidence to support that it may act as an endocrine disrupting chemical. Juvenile rainbow trout (Oncorhynchus mykiss) were caged at 4 sites in Oshawa Creek during the fall and spring of 2008 and 2009 while another group was intra-peritoneal injected in the laboratory with varying concentrations (ppm) of endosulfan. Plasma vitellogenin (VTG) levels, liver ethoxyresorufin-O-deethylase (EROD), citrate synthase (CS), lactate dehydrogenase (LDH), and brain acetylcholine esterase (AChE) (caged fish only) enzymatic activities were measured. Trout injected with endosulfan experienced an increase of the anaerobic (LDH activity) and a decrease of the aerobic (CS activity) metabolic pathways, while male VTG levels increased. Since it was a singular injection, VTG results have to be confirmed. Fall caged trout showed increased EROD activity and inhibited AChE activity while those caged in the spring experienced an unexpected exposure to the lampricide 3-Trifluoro-Methyl-4-Nitro-Phenol (TFM) which disrupted metabolic parameters (inhibited CS and increased LDH activity). Both fall and spring caged trout experienced no induction of VTG activity. Further research is needed since the spring exposure was altered due to the unplanned TFM treatment and thus did not represent a valid temporal replicate.

Previous work (Thorgaard, G. H. et al., Aquatic Toxicology 46:121-126, 1999) showed triploid rainbow trout (0. mykiss) given embryonic carcinogen bath exposures had significant reduction of induced tumors relative to diploids. In the present study, trout were made triploid by thermal shock after fertilization. At age of 5 months they were given dietary carcinogen: aflatoxin B1 (AFB₁) or 7,12-dimethylbenz[a]anthracene (DMBA) for 30 or 120 days. The dietary exposures were at known tumorigenic levels (100, 200 and 300 ppb AFB₁; 250, 500 and 850 ppm DMBA). At about 16 months after fertilization the fish were sacrificed and tumor incidence and multiplicity were assessed. At all levels of carcinogen and in all tumorous organs tumor incidence was lower in the triploid fish. For DMBA-fed fish it was seen that the diploid:triploid incidence ratios ranged from 2.0 to 9.0 and for AFB₁ from 3.1 to 6.0. Weight class analyses dissociated the tumor incidence effects of growth from the effects of triploidy. Weight classes plotted against logit tumor incidence at all doses and durations showed parallel logistic lines. In every case the triploid curve was substantially lower than the diploid curve, showing the independent suppressive effect of triploidy. Fifteen triploid DMBA liver tumors were examined by direct cycle-sequencing of p53 PCR products across the exons 5, 7 and 8 known to contain nearly all human tumor p53 mutations. There were no p53 mutations seen at, or above, the present threshold of detection, (for radiolabeled manual sequencing, under 5% of mutant in normal). Fluorescent sequencing of 15 stomach tumors, also showed no p53 mutations in the hotspot-containing exons. Mutation detection by sequencing the trout Ki-ras1 gene, ortholog human KRAS2, showed codons 12, and 61 mutations in DMBA-fed trout liver and stomach tumors. The DMBA liver tumor Ki-ras1 mutation incidence showed no change by ploidy. There was a significant reduction in Ki-ras1 exon 1 mutations in triploid stomach tumors (5% in triploids v. 33% in diploids, Fisher's Exact test p<O.O5). AFB₁ liver tumors showed Ki-ras1 mutation incidence of 75% (9/12) in diploids and 90% (9/10) triploids, nearly all in exon 1, this mutation difference with respect to ploidy did not reach significance. / Graduation date: 2001