Thesis (Ph. D.)--University of Washington, 2000. / Vita. Includes bibliographical references (leaves 70-78).

Thesis (Ph. D.)--University of Washington, 2003. / Vita. Includes bibliographical references (leaves 110-117).

Thesis (Ph. D.)--University of Washington, 2002. / Vita. Includes bibliographical references (leaves 130-160).

The efficacy of interventions to control HIV spread depends upon many features of the communities where they are implemented, including not only prevalence, incidence, and per contact risk of transmission, but also properties of the sexual or transmission network. For this reason, HIV epidemic models have to take into account network properties including degree distribution and mixing patterns. The use of sampled data to estimate properties of a network is a common practice; however, current network generation methods do not account for the uncertainty in the estimates due to sampling. In chapter 1, we present a framework for constructing collections of networks using sampled data collected from ego-centric surveys. The constructed networks not only target estimates for density, degree distributions and mixing frequencies, but also incorporate the uncertainty due to sampling. Our method is applied to the National Longitudinal Study of Adolescent Health and considers two sampling procedures. We demonstrate how a collection of constructed networks using the proposed methods are useful in investigating variation in unobserved network topology, and therefore also insightful for studying processes that operate on networks. In chapter 2, we focus on the degree to which impact of concurrency on HIV incidence in a community may be overshadowed by differences in unobserved, but local, network properties. Our results demonstrate that even after controlling for cumulative ego-centric properties, i.e. degree distribution and concurrency, other network properties, which include degree mixing and clustering, can be very influential on the size of the potential epidemic. In chapter 3, we demonstrate the need to incorporate information about degree mixing patterns in such modeling. We present a procedure to construct collections of bipartite networks, given point estimates for degree distribution, that either makes use of information on the degree mixing matrix or assumes that no such information is available. These methods permit a demonstration of the differences between these two network collections, even when degree sequence is fixed. Methods are also developed to estimate degree mixing patterns, given a point estimate for the degree distribution.

Most antibiotics, including the drugs currently used for treating tuberculosis (TB), were first discovered as molecules that inhibit bacterial growth in laboratory culture conditions and later translated to infection models and clinical use. Mycobacterium tuberculosis (Mtb) has evolved specifically to survive in its human host, and it is in this infectious context that new drugs need to work. The host environment is characterized by a multitude of antimicrobial defenses induced by the immune system, and we can leverage these defenses to kill Mtb in vivo. Mtb employs a diverse set of responses to survive host defenses. By blocking these responses, we can make Mtb more susceptible to host immunity, turning these previously impotent defenses into effective strategies of immune control.

The Bayesian Markov Chain Monte Carlo(MCMC) approach to inference is commonly used to estimate the parameters in spatial infectious disease models. However, such MCMC analyses can pose a hefty computational burden. Here we present new method to reduce the computing time cost in such MCMC analyses and study its usefulness. This method is based a round the discretization of the spatial parameters in the infectious disease model. A normal approximation of the posterior density of the output from the original model will be compared to that of the modified model, using the Kullback-Leibler(KL) divergence measure.

Porcine Periweaning Failure-to-Thrive syndrome (PFTS) is a clinical syndrome of newly weaned pigs with unknown etiology and characterized by anorexia, lethargy and progressive debilitation. The hypothesis of this thesis is that PFTS is an infectious disease. Investigation in an index farm affected by PFTS from Saskatchewan Canada ruled out most common swine pathogens as the etiology and identified several lesions that were consistent across many cases. A larger study including multiple farms in North America was then undertaken. A total of 8 farms were investigated, within which 5 met the clinical definition of PFTS. Gross and histological examinations were performed on 8 case and 4 control pigs on each farm. Detection of relevant porcine pathogens, complete blood count, serum chemistry, and serum cytokine analysis were performed on each pig. Thymic atrophy, superficial gastritis and small intestinal villous atrophy were significantly more prevalent in case pigs compared to control pigs. All case pigs had at least two of these three lesions. All case and control pigs were negative for porcine reproductive and respiratory syndrome virus, swine influenza virus and were free of porcine circovirus associated diseases. Although several pathogens, such as porcine cytomegalovirus, haemagglutinating encephalomyelitis virus, porcine enteric calicivirus, group A rotavirus, enteroviruses and Cystoisospora suis were detected in some of the case and control pigs, none were associated with clinical status. Clinical pathology findings of case pigs was consistent with anorexia and dehydration, such as increases in haematocrit, blood urea, serum bilirubin, albumin, beta-hydroxybutyrate and decreases in blood glucose, calcium and phosphorous. Case pigs had similar levels to IL1-β than control pigs, which suggested that PFTS was not a result of excessive cytokines. In subsequent experiments, a snatched-farrowed porcine-colostrum-deprived (SF-pCD) pig model was developed and tissue homogenates were used to inoculate SF-pCD pigs in an attempt to reproduce the clinical signs of PFTS. The SF-pCD pigs were immunologically characterized and shown to be suitable for inoculation studies. However, inoculation of tissue homogenate from PFTS pigs failed to reproduce the clinical signs of PFTS in SF-pCD pigs. All together, PFTS is a clinical syndrome with consistent pathological and serum analytical changes among affected pigs. Despite the efforts of this research to establish an infectious etiology, there is a lack of evidence that PFTS is an infectious disease.

Equine influenza virus causes a mild, self-limiting upper respiratory disease in its natural host. In stark contrast, equine influenza viruses of the H7N7 subtype produce lethal infection in BALB/c mice. This dissertation explored the mechanism underlying the differential pathogenicity of the equine H7N7 influenza virus observed in horses and BALB/c mice. Initially, a comparative study of the pathogenesis was conducted in BALB/c mice inoculated intranasally with a representative isolate of either H7N7 or H3N8 subtype equine influenza virus. All H3N8 virus-infected mice survived the infection whereas 100% mortality was documented for the mice receiving the H7N7 virus by day 8 post infection. Both viruses replicated to a similar degree in the lungs at the early stages of infection. However, after day 2 post infection until the death of the mice, the pulmonary viral loads of the H7N7 group were significantly higher than those of the control, whereas the H3N8 virus was eventually eradicated from the mice at day 7 p.i. Correspondingly, a vigorous pro-inflammatory cytokine response in the lung was induced by the H7N7 virus but not the H3N8 virus, which reflected a desperate attempt by the host immune responses to restrain the overwhelming infection. The H7N7 virus was poorly sensitive to the innate immune containment, resulting in a significant higher cumulative mortality rate than that of the control virus in chicken embryos aged 9 days and older. On the contrary, in horses, replication of the paired viruses was completely cleared by the host immune responses at day 7 p.i. and the infections produced an acute yet non-lethal illness, albeit the H3N8 virus induced generally more pronounced clinical manifestations than the H7N7 virus. The clinical severity correlated to the difference in cytokine-inducing capacity between the two viruses in horses, as evidenced by the finding that the H3N8 virus triggered significantly higher levels of gene transcription of multiple key inflammatory cytokines in the circulation than those seen for the H7N7 virus. In addition, equine peripheral monocyte-derived macrophages were found to be a target of equine influenza virus and can support the productive replication of the virus in vitro.

The mammalian immune system is equipped to both eliminate pathogenic microorganisms and tumors, while remaining in homeostasis with commensal species at mucosal surfaces and tolerant towards self. Suppressor regulatory T cells (Tregs) are a major sentinel of this immunological tolerance. Induced Tregs (iTregs) arise in the periphery following the integration of cues from the metabolites, cytokines, etc. which make up its milieu. Dysregulation of iTreg development, function or homing underlies the etiology of many autoimmune diseases and immunopathologies. The amelioration or prevention of multiple murine disease models by boosting Treg cell numbers foreshadows clinical efficacy of iTreg therapy, but an incomplete understanding of Treg development has thus far prevented successful translation. Therefore, we considered the basic biology of T cell fate decision making from two unique, but integrated angles. First, we show that the stimulation of PPARγ in human T cells upregulates RDH10, a molecule which catalyzes the rate limiting step in the oxidation of retinol to transcriptionally active all-trans retinoic acid (ATRA), a positive regulator of iTreg development. This functionally intact pathway endows T cells the ability to autonomously sense and respond to retinoid signals present during Treg development and at tissue sites. Next, we asked questions about how T cells sense nutrient and oxygen availability as they differentiate. Tregs lacking the serine/threonine kinase PINK1 have limited activation-induced phosphorylation of Akt and oxidative phosphorylation rates, and reduced suppressor function. Notably, the uncoupling of iTreg function from normal FoxP3 expression reinforces the recent hypothesis that the PI3K/Akt/mTORC1 axis and metabolic checkpoints are decisive players in the acquisition of suppressor activity. Ultimately, the studies described herein converge on Akt and metabolism, and contribute to our understanding of how T cells integrate diverse signals present during fate determinism, provoking future Treg based therapeutics.

This thesis presents the development and optimization of a high-throughput fluorescence microbead based approach for multiplexed, large scale medical diagnostics of biological fluids. Specifically, different sizes of semiconductor nanocrystals, called quantum dots, are infused into polystyrene microspheres, yielding a set of spectrally unique optical barcodes. The surface of these barcodes are then used for sandwich assays with target molecules and fluorophore-conjugated detection antibodies, changing the optical spectra of beads that have associated with (or captured) biomolecular targets. These assayed microbeads are analyzed at a single bead level in a high-throughput manner using an electrokinetic microfluidic system and laser induced fluorescence. Optical signals collected by solid state photodetectors are then processed using novel signal processing algorithms. This document will discuss developments made in each area of the platform as well as optimization of the platform for improved future performance.