Many inflammatory diseases are growing increasing common in the aging society of Taiwan. Inflammation cascades can cause diseases such as rheumatoid arthritis, osteoarthritis, chronic asthma, multiple sclerosis, and so on. The clinically used anti-inflammatory drugs have many side effects and are expensive. Therefore, it is imperative that we find alternatives to these drugs. Marine natural compounds offer great hope in the development of drugs for treating inflammatory diseases. In the present study, we found that Chao-10, which is a marine-derived compound isolated from Formosan soft coral, significantly inhibited the expression of the pro-inflammatory protein, inducible nitric oxide synthase (iNOS), in the lipopolysaccharides (LPS)-stimulated RAW 264.7 macrophage cell line. We suggest that Chao-10 may serve as a potential new anti-inflammatory agent. However, the mechanism by which the anti-inflammatory effects of Chao-10 are mediated is yet unclear. Therefore, we performed two-dimensional electrophoresis (2-DE) to investigate the regulatory mechanism for the anti-inflammatory effect of Chao-10. We isolated some proteins that may be involved in the anti-inflammatory mechanism of Chao-10. In addition, we used immunoprecipitation to find that nucleophosmin (NPM) could interact with nuclear factor kappa B (NF-£eB). Therefore, we hypothesize that nucleophosminmay be involved in the regulation of NF-£eB to enhance the down-regulation of iNOS proteins. In summary, the anti-inflammatory effects of Chao-10 are probably mediated through the some other signaling pathway. Importantly, Chao-10 not only offers some new biomarkers of inflammation but also provides an encouraging outlook on therapeutic approaches.

Chronic kidney disease (CKD) is a major cause of illness in dogs, and it is commonly caused by glomerular diseases that result in proteinuria and a progressive decline in renal function. Despite the importance of glomerular lesions, tubulointerstitial fibrosis identified by histologic evaluation of renal biopsies correlates best with renal function. However, performing a renal biopsy is invasive. Most current non-invasive tests for renal function lack adequate sensitivity and specificity for renal disease. Proteinuria can be both a sensitive and specific marker for renal damage. However, its evaluation in veterinary medicine beyond determination of the magnitude of proteinuria (e.g., urine protein:creatinine ratio (UPC)) is limited. Therefore, in this report, further evaluation of the UPC was performed to aid in the monitoring of renal disease progression and response to treatment. In addition, qualitative evaluation of proteinuria was performed in dogs with progressive CKD in order to identify better non-invasive markers for tubulointerstitial injury. The day-to-day variability of the UPC was determined utilizing data obtained from female dogs that are carriers for X-linked hereditary nephropathy (XLHN). Despite an unchanging magnitude of proteinuria in these dogs, substantial variation in their UPC was observed. Using these results, guidelines were suggested to help assess whether disease progression or treatment leads to a significant change in UPC. Qualitative characterization of proteinuria in dogs with CKD was performed using urine from male dogs affected with XLHN, and results were correlated with clinical and histologic findings concerning renal function and damage. The two discovery proteomic techniques utilized (chromatographic chip array and two-dimensional gel electrophoresis) revealed several proteins that have not previously been implicated as markers for canine CKD, providing a basis for future studies. Specific assays for urinary biomarkers of renal injury were used to serially evaluate renal function in these dogs. All proteins evaluated proved to be sensitive markers for renal damage. However, only retinol binding protein provided clear evidence for renal disease progression. These results will provide the foundation for future studies aimed at monitoring urinary biomarkers in dogs with CKD, which will ultimately help veterinarians better diagnose and monitor proteinuric renal disease.

The small-sample size issue is a prevalent problem in Genomics and Proteomics today. Bootstrap, a resampling method which aims at increasing the efficiency of data usage, is considered to be an effort to overcome the problem of limited sample size. This dissertation studies the application of bootstrap to two problems of supervised learning with small sample data: estimation of the misclassification error of Gaussian discriminant analysis, and the bagging ensemble classification method. Estimating the misclassification error of discriminant analysis is a classical problem in pattern recognition and has many important applications in biomedical research. Bootstrap error estimation has been shown empirically to be one of the best estimation methods in terms of root mean squared error. In the first part of this work, we conduct a detailed analytical study of bootstrap error estimation for the Linear Discriminant Analysis (LDA) classification rule under Gaussian populations. We derive the exact formulas of the first and the second moment of the zero bootstrap and the convex bootstrap estimators, as well as their cross moments with the resubstitution estimator and the true error. Based on these results, we obtain the exact formulas of the bias, the variance, and the root mean squared error of the deviation from the true error of these bootstrap estimators. This includes the moments of the popular .632 bootstrap estimator. Moreover, we obtain the optimal weight for unbiased and minimum-RMS convex bootstrap estimators. In the univariate case, all the expressions involve Gaussian distributions, whereas in the multivariate case, the results are written in terms of bivariate doubly non-central F distributions. In the second part of this work, we conduct an extensive empirical investigation of bagging, which is an application of bootstrap to ensemble classification. We investigate the performance of bagging in the classification of small-sample gene-expression data and protein-abundance mass spectrometry data, as well as the accuracy of small-sample error estimation with this ensemble classification rule. We observed that, under t-test and RELIEF filter-based feature selection, bagging generally does a good job of improving the performance of unstable, overtting classifiers, such as CART decision trees and neural networks, but that improvement was not sufficient to beat the performance of single stable, non-overtting classifiers, such as diagonal and plain linear discriminant analysis, or 3-nearest neighbors. Furthermore, the ensemble method did not improve the performance of these stable classifiers significantly. We give an explicit definition of the out-of-bag estimator that is intended to remove estimator bias, by formulating carefully how the error count is normalized, and investigate the performance of error estimation for bagging of common classification rules, including LDA, 3NN, and CART, applied on both synthetic and real patient data, corresponding to the use of common error estimators such as resubstitution, leave-one-out, cross-validation, basic bootstrap, bootstrap 632, bootstrap 632 plus, bolstering, semi-bolstering, in addition to the out-of-bag estimator. The results from the numerical experiments indicated that the performance of the out-of-bag estimator is very similar to that of leave-one-out; in particular, the out-of-bag estimator is slightly pessimistically biased. The performance of the other estimators is consistent with their performance with the corresponding single classifiers, as reported in other studies. The results of this work are expected to provide helpful guidance to practitioners who are interested in applying the bootstrap in supervised learning applications.

The overall goal of this research was to investigate oxidative stress related changes to the proteomes of 3T3-L1 adipocytes and an inducible CYP2E1 expressing HepG2 cells. Enhanced oxidative stress in hypertrophic adipocytes is associated with metabolic dysregulation and insulin resistance. Because mitochondria generate reactive oxygen species (ROS), we monitored changes to the adipocyte mitochondrial proteome during differentiation and enlargement. We labeled mitochondrial extracts from 3T3-L1 cells that were 0, 4, 7, 10, 14, and 18 days post differentiation with iTRAQ, followed by MS based identification. We found citric acid cycle proteins such as pyruvate carboxylase, citrate synthase, as well as beta-oxidation enzymes; cartinine acyl transferase and long-chain enoyl-CoA hydratase up-regulated from 7 through 18 days post differentiation onset. These data indicate TCA up-regulation for enhanced metabolic and citrate output necessary for lipid synthesis in adipocytes. Paradoxically, the data also show the simultaneous increase in the fatty acid oxidation, indicating a metabolic overdrive state. Biochemical assays showing peaks in ATP and ROS generation in 3 day old adipocytes provide further evidence of this overdrive state. A second peak in ROS generation occurred in 10 day old adipocytes; concurrent ATP generation reduced to near pre-adipocyte levels and this may indicate a metabolic shift that may be responsible for increased oxidative stress in hypertrophic adipocytes. We developed a doxycycline inducible CYP2E1 expressing HepG2 cell line using the pTet-On/pRevTRE expression system to allow greater control and sensitivity in the generation CYP2E1 mediated oxidative stress. Our cell line (RD12) demonstrated stability and tight expression control. After induction, RD12 cells showed 30 percent higher CYP2E1 activity when compared to the constitutive E47 cell line. RD12 cells showed 30 percent greater toxicity than E47 cells and 25 percent less free glutathione when exposed to 20 mM acetaminophen, indicating RD12 cells are more sensitive to the effects reactive intermediates and oxidative stress generated by CYP2E1. We conducted a survey of the toxicity of dietary fatty acids (oleic, linoleic, and palmitic) on HepG2 cells to determine fatty acid doses that induced metabolic changes, but did not cause excessive cell death. The dose of 0.20 mM linoleic and palmitic acid for 48 hours produced low toxicity, but oleic acid actually produced lower toxicity than untreated cells. After exposure cells were treated with a pro-oxidant to determine which fatty acid increased the susceptibility to protein carbonylation. The carbonylated protein isolation procedure indicated the palmitic acid may induce more carbonylation than oleic acid, but greater efficiency in the isolation procedure is required for a confidant determination.

Bacteria cause significant morbidity and mortality throughout the world, including deadly diseases such as tuberculosis, meningitis, cholera, and pneumonia. Timely and accurate bacterial identification is critical in areas such as clinical diagnostics, environmental monitoring, food safety, water and air quality assessment, and identification of biological threat agents. At present, there is an established need for high throughput, sensitive, selective, and rapid methods for the detection of pathogenic bacteria, as existing methods, while nominally effective, have failed to sufficiently reduce the massive impact of bacterial contamination and infection. The work presented in this thesis focuses on addressing this need and augmenting conventional microorganism research through development of mass spectrometry (MS)-based proteomic applications. MS, a well established tool for addressing biological problems, offers a broad range of laboratory procedures that can be used for taxonomic classification and identification of microorganisms. These methods provide a powerful complement to many of the widely used molecular biology approaches and play critical functions in various fields of science. While implementation of modern biomolecule-identifying instrumentation, such as MS, has long been postulated to have a role in the microbiology laboratory, it has yet to be accepted on a large scale. Described in this document are MS methods that erect strong foundations on which new bacterial diagnostics may be based. A general introduction on key aspects of this work is presented in Chapter 1, where different approaches for detection of pathogenic bacteria are reviewed, and an overview regarding MS and microbial identification is provided. Chapter 2 presents the first implementation of microbial identification via rapid, open air Direct Analysis in Real Time MS (DART MS) to generate ions directly from microbial samples, including the disease-causing bacteria, Coxiella burnetii, Streptococcus pyogenes, and Escherichia coli. Chapter 3 expands on whole cell C. burnetii MS analysis and presents a rapid differentiation method to the strain-level for C. burnetii using mass profiling/fingerprinting matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS and multivariate pattern recognition. Chapter 4 presents a unique "top-down" proteomics approach using 15N-labeled bacteriophage amplification coupled with MALDI-TOF MS as a detector for the rapid and selective identification of Staphylococcus aureus. Chapter 5 extends the idea of using isotopically labeled bacteriophage amplification by implementing a "bottom-up" proteomics approach that not only identifies S. aureus in a sample, but also quantifies the bacterial concentration in the sample using liquid chromatography-electrospray ionization tandem mass spectrometry (LC-ESI/MS/MS) as a detector. In conclusion, Chapter 6, summarizes and contextualizes the work presented in this dissertation, and outlines how future research can build upon the experimentation detailed in this document.

During vertebrate gastrulation, a well-orchestrated series of cell movements leads to the formation of the three germ layers: ectoderm, mesoderm and endoderm. In zebrafish, a model organism for vertebrate development, the mesendodermal progenitor cells separate from the ectodermal cells and migrate towards the animal pole. To identify proteins controlling these processes, I used a comparative proteomic approach following two alternative strategies: (1) Based on the notion that Wnt11 regulates cell movement and morphology during gastrulation independent of transcriptional regulation, I performed a screen aimed at the identification of proteins phosphorylated upon Wnt11 signalling. To regulate Wnt11 expression tightly, I engineered a transgenic slb/wnt11-/- fish line expressing wnt11 under the control of a heat shock promoter. Using this line, I performed a quantitative comparison of protein phosphorylation with or without Wnt11 pathway activation by analysing 32P-labelled embryo extracts on 2D gels. (2) Since these experiments did not reveal any Wnt11 targets, I addressed, in the second approach, proteomic differences causal for the changes in cell adhesion and motility observed in mesendodermal cells upon involution. Quantitative 2D gel analysis comparing ectodermal and mesendodermal cells revealed 37 significantly regulated spots, 36 of which I identified by mass spectrometry. Interestingly, the majority of these proteins were not regulated on a transcriptional level as determined by an accompanying microarray analysis confirming the complementary nature of proteomics and transcriptomics. Among the identified targets, several proteins, including Ezrin2, had previously been assigned a cytoskeleton-related function. I characterised Ezrin2 in more detail showing that Ezrin2 is specifically activated by phosphorylation in mesendodermal cells and that it is required for proper gastrulation movements. In the course of this study, I developed techniques for proteomic analysis of early zebrafish embryos, including a protocol to remove the yolk. I identified several cytoskeleton-related proteins in a comparative proteomic screen for regulators of gastrulation movements. The subsequent characterisation of Ezrin2 confirmed the power of proteomics for the analysis of developmental processes. In conclusion, this work provides a foundation to study developmental and cell biological questions in early zebrafish embryos using proteomics.