Heart failure (HF) following myocardial infarction (MI) is the leading cause of mortality and morbidity worldwide. Existing medical and interventional therapies can only reduce the cardiomyocytes (CMs) lost during MI. They are unable to replenish the permanent loss of CMs and this contributes to progressive pathological left ventricular (LV) remodeling and HF. Cell-based therapies using adult stem cells or embryonic stem cells (ESCs) and their cardiac derivatives have frequently been explored as a potential therapeutic approach to restore cardiac function in HF. The objectives of this thesis are to evaluate the efficacy and safety of different approaches of stem cell based therapy to improve cardiac function using small and large animal MI models. In Chapter 3, we studied the functional consequences of direct intramyocardial transplantation of ESCs and ESC-derived cardiomyocytes (ESC-CMs) in a murine model of acute MI. LV ejection fraction (LVEF) and maximal positive or negative pressure derivative (dP/dt) improved 4 weeks after transplantation of either ESCs or ESC-CMs. Nevertheless there was a higher incidence of inducible ventricular tachyarrhythmia (VT) and higher mortality in animals transplanted with ESC-CMs than those with ESCs. At a single cell level, ESC-CMs exhibited immature electrophysiological properties such as depolarized resting membrane potential (RMP), longer action potential duration (APD) and automaticity. In Chapter 4, we tested the hypothesis that genetic modification of these immature electrophysiological properties of ESC-CMs by overexpression of Kir2.1 gene encoding the ion channels for IK1, may alleviate the pro-arrhythmic risk. In this study, Kir2.1 channels expression could be controlled with the administration of doxycycline (DOX). The DOX-treated ESC-CMs were more mature with hyperpolarized RMP and shorter APD than their counterparts without DOX treatment. A similar improvement in LV systolic function was observed 4 weeks after both DOX treated and untreated ESCCMs transplantation, although those animals transplanted with DOX-treated ESC-CMs had a significantly lower incidence of spontaneous and inducible VT. Histological analysis in both studies suggested that the major mechanisms of improvement in cardiac function were related to angiogenesis and low apoptosis rate of native cardiomyocytes mediated via paracrine effects. Importantly, very limited retention of ESC-CMs was observed 4 weeks after transplantation. Cell-based patches that use different bioengineering techniques have been proposed to improve cell retention and survival following transplantation. In Chapter 5, the efficacy of a passive epicardial patch was tested in a chronic large animal MI model with HF created with catheter-based coronary embolization. The implantation of an epicardical patch over the infarcted LV region was performed 8 weeks after MI in pigs with impaired LVEF. At week 20, pigs implanted with epicardical patches had significantly thicker LV wall thickness at the infarction sites, smaller LV dilation and better LV systolic function compared with control animals. The expression of MMP-9 was significant lower in the epicardical patch group at the peri-infarct zones. These findings suggested that a passive epicardial patch can improve LV function in HF and provides important proof-of-principle data to support its use as a platform for delivery of cell-based therapies after MI. / published_or_final_version / Medicine / Doctoral / Doctor of Philosophy

SWI/SNF is a chromatin-remodeling complex and Brahma (BRM) is the ATPase subunit of SWI/SNF. BRM regulates transcription by remodeling the nucleosomes at promoter regions. BRM is also associated with RNA and affects pre-mRNA processing together with other SWI/SNF subunits. In this thesis, I will discuss the roles of BRM in both transcription and pre-mRNA processing. In Paper I, we showed that BRM, as well as other SWI/SNF subunits SNR1 and MOR, affects the alternative processing of a subset of pre-mRNAs, as shown by microarray analysis. This observation was validated by RNAi experiments both in Drosophila S2 cells and in vivo. In Paper II, we characterized the trans-splicing of transcripts derived from the mod(mdg4) gene. RNA interference (RNAi) and overexpression experiments revealed that BRM regulates the trans-splicing of mod(mdg4)-RX in an ATPase independent manner. In Paper III, we analyzed the expression of two BRM-target genes identified in Paper I, CG44250 and CG44251. RNAi and overexpression experiments showed that the expression levels of these two genes were affected by BRM in a manner that is independent of its ATPase activity. Transcriptome analysis further proved that BRM affects gene expression both in ATPase dependent and independent manners. In Paper IV, we showed that BRM is present at the 3’-end of two analyzed genes, CG5174 and CG2051. BRM facilitates the recruitment of the cleavage and polyadenylation machinery to the cleavage sites through protein-protein interactions that do not require the ATPase activity of BRM. Morevoer, BRM promotes the cleavage of the CG5174 and CG2051 pre-mRNAs. To sum up, SWI/SNF plays important roles not only in transcription but also in pre-mRNA processing. To regulate transcription, BRM can either act as an ATPase-dependent chromatin remodeler or in a manner that does not involve ATPase activity. Additionally, BRM interacts with RNA-binding proteins to regulate the processing of a subset of pre-mRNAs, and this function of BRM is independent of its chromatin remodeling activity. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Manuscript. Paper 4: Manuscript.</p>

published_or_final_version / Architecture / Master / Master of Architecture

The purpose of this dissertation is to establish a unified theory of porohyperelasticity with transport and growth and to demonstrate the capability of this theory using a finite element model developed in MATLAB. The theory of volumetric growth is combined with the theory of mixed porohyperelasticity with transport and swelling (MPHETS) to derive a new method that models growth of biological soft tissues. The conservation equations and constitutive equations are developed for both solid-only growth and solid-fluid growth. An axisymmetric finite element framework is introduced for the new theory of growing MPHETS (GMPHETS). To demonstrate the difference of the GMPHETS model from a traditional hyperelastic (HE) growth model, several finite element test problems with example growth laws are considered, including time-dependent, concentration-dependent, and stress-dependent growth. In particular, this work demonstrates that the solid-only growth of an MPHETS model of a stylized artery results in a more uniform hoop stress than in a HE model under solid-only growth for the same amount of growth time using the same growth law. This may have implications in the context of developing residual stresses in soft tissues under intraluminal pressure. To my knowledge, this is the first description of an MPHETS model with growth. The developed computational framework can be used together with novel in-vitro and in-vivo experimental approaches to identify the governing growth laws for various soft tissues.

The thesis of this dissertation is that automating domain remodeling in AI planning using macro operators and making remodeling more flexible and applicable can improve the planning performance and can enrich planning. In this dissertation, we present three novel ideas: (1) we present an instance-specific domain remodeling framework, (2) we recast the planning domain remodeling with macros as a parameter optimization problem, and (3) we combine two domain remodeling approaches in the instance-specific remodeling context. In the instance-specific domain remodeling, we choose the best macro-augmented domain model for every incoming problem instance using a predictor that relies on previously solved problem instances to estimate the macros to be added the domain. Training the predictor is achieved off-line based on the observed relation between the instance features and the planner performance in the macro-augmented domain models. On-line, the predictor is used to find the best remodeling of the domain based on the problem instance features. Our empirical results over a number of standard benchmark planning domains demonstrate that our predictors can speed up the fixed-remodeling method that chooses the best set of macros by up to 2.5 times. The results also show that there is a large room for improving the performance using instance-specific over fixed remodeling approaches. The second idea is recasting the domain remodeling with macros as a parameter optimization. We show that this remodeling approach can outperform standard macro learning tools, and that it can significantly speed up the domain evaluation preprocessing required to train the predictors in instance-specific remodeling, while maintaining similar performance. The final idea applies macro addition and operator removal to the instance-specific domain remodeling. While maintaining an acceptable probability of solubility preservation, we build a predictor that adds macros and removes original operators based on the instance’s features. The results show that this new remodeling significantly outperforms the macro-only fixed remodeling, and that it is better than the fixed domain models in a number of domains.

The thesis of this dissertation is that automating domain remodeling in AI planning using macro operators and making remodeling more flexible and applicable can improve the planning performance and can enrich planning. In this dissertation, we present three novel ideas: (1) we present an instance-specific domain remodeling framework, (2) we recast the planning domain remodeling with macros as a parameter optimization problem, and (3) we combine two domain remodeling approaches in the instance-specific remodeling context. In the instance-specific domain remodeling, we choose the best macro-augmented domain model for every incoming problem instance using a predictor that relies on previously solved problem instances to estimate the macros to be added the domain. Training the predictor is achieved off-line based on the observed relation between the instance features and the planner performance in the macro-augmented domain models. On-line, the predictor is used to find the best remodeling of the domain based on the problem instance features. Our empirical results over a number of standard benchmark planning domains demonstrate that our predictors can speed up the fixed-remodeling method that chooses the best set of macros by up to 2.5 times. The results also show that there is a large room for improving the performance using instance-specific over fixed remodeling approaches. The second idea is recasting the domain remodeling with macros as a parameter optimization. We show that this remodeling approach can outperform standard macro learning tools, and that it can significantly speed up the domain evaluation preprocessing required to train the predictors in instance-specific remodeling, while maintaining similar performance. The final idea applies macro addition and operator removal to the instance-specific domain remodeling. While maintaining an acceptable probability of solubility preservation, we build a predictor that adds macros and removes original operators based on the instance’s features. The results show that this new remodeling significantly outperforms the macro-only fixed remodeling, and that it is better than the fixed domain models in a number of domains.

No description available.

No description available.

No description available.

No description available.