published_or_final_version / Zoology / Doctoral / Doctor of Philosophy

published_or_final_version / abstract / Biochemistry / Doctoral / Doctor of Philosophy

published_or_final_version / Biological Sciences / Master / Master of Philosophy

PRDM16 is a transcriptional co-regulator that is highly expressed in HSCs and required for their maintenance. It is also involved in translocations in acute myeloid leukemia (AML), myelodysplastic syndromes (MDS) and T-cell acute lymphoblastic leukemia. Prdm16 is expressed as both full-length (f Prdm16) and short-length (s-Prdm16) isoforms, the latter lacking an N-terminal PR domain homologous to SET methyltransferase domains. The roles of both isoforms in normal and malignant hematopoiesis are unclear. In chromosomal rearrangements involving PRDM16, the PR domain is deleted. Furthermore, overexpression of s-Prdm16, but not f-Prdm16, can cause leukemia in a p53-/- background predisposed to malignancy. Based on this, s-Prdm16 has been proposed as an oncogene whereas f-Prdm16 has been suggested to possess tumor suppressor activity. The aim of this thesis was to more clearly elucidate the role of each Prdm16 isoform in normal and malignant hematopoiesis. We first showed that Prdm16 is essential for adult HSC maintenance using a conditional deletion mouse model specific for hematopoietic cells, as previous findings using an embryonic-lethal global Prdm16-/- mouse demonstrated this only in fetal liver. We then found, using a specific f-Prdm16-/- mouse model, that full-length Prdm16 is essential for HSC maintenance and induces multiple genes involved in GTPase signaling and represses inflammation. Based on a comparison of Prdm16-/- HSCs lacking both isoforms, and f-Prdm16-/- HSCs which express s-Prdm16, we were able to infer some hematopoietic properties of s-Prdm16 – namely that this isoform induces inflammatory gene expression and supports development of a Lineage-Sca1+cKit- lymphoid progenitor distinct from CLPs which predominantly differentiates into marginal zone B cells. s-Prdm16 expression alone, however, was not sufficient to maintain HSCs. We used a mouse model of human MLL-AF9 leukemia and found that leukemia derived from Prdm16-deficient HSCs had extended latency, although expression of Prdm16 decreases during MLL-AF9 transformation and is undetectable in ex vivo leukemic cells. Forced expression of f-Prdm16 in these cells further extended leukemic latency, while forced expression of s-Prdm16 shortened latency. Gene expression profiling using RNAseq indicated that forced expression of f-Prdm16 resulted in altered respiratory metabolism of MLL-AF9 cells, whereas expression of s-Prdm16 induced a strong inflammatory gene signature, comparable to that seen in HSCs expressing only s-Prdm16. Several inflammatory cytokines and chemokines induced by s-Prdm16 are associated with MDS and with a worse prognosis in human AML. Furthermore, leukemia expressing s-Prdm16 had an elevated number of cells with abnormal nuclei, characteristic of dysplasia. Finally, we performed an analysis of PRDM16 in human AML from the publically-available Cancer Genome Atlas dataset, containing clinical and gene expression data for 179 cases of AML. PRDM16 expression negatively correlated with overall survival, both in the entire dataset and in the NPM1 mutated and MLL¬-rearranged subsets, and s-PRDM16 exhibited a stronger correlation than f-PRDM16. HOX gene expression correlated with PRDM16 expression, suggesting that HOX genes may positively regulate PRDM16 expression in AML. In NPM1-mutant and MLL-rearranged subsets of AML, we also found that high PRDM16 expression correlated with an inflammatory gene signature, thus corroborating our findings in mouse MLL-AF9. Our findings demonstrate distinct roles for Prdm16 isoforms in both normal hematopoiesis and AML, and identify s-Prdm16 as one of the drivers of prognostically-adverse inflammatory gene expression in leukemia.

Herpes simplex virus type 1 (HSV-1) is a virus commonly causing cold sores in humans, however, virulent infections are known to produce debilitating encephalitis and death. HSV-1 transcription is carried out by the host cell RNA polymerase II in a tightly regulated temporal cascade. The first genes transcribed, the a genes, are activated in the absence of viral DNA synthesis. Transcription of the other temporal classes, the β, βγ, and γ genes is dependent upon the protein products of the a genes for activation. The purpose of this study was to investigate the factors that contribute to this rigid regulation of HSV-1 transcription. This investigation sought to identify some of the cellular and viral transcription factors that activate transcription of genes of the later kinetic classes. Two separate approaches were utilized in these investigations. 1) In vitro transcription using a soluble, cell free system to study the transcriptional regulation of the VP5 gene, and 2) DNA competition binding assays to identify and characterize the protein-DNA complexes resulting from interaction between the cisacting DNA sequences of the VP5 gene, other viral genes, and the proteins that bind to them. Attempts at in vitro transcription of β, βγ, and γ genes were unsuccessful. Because these genes require a products for activation, it was necessary to prepare nuclear extracts from infected cells. However, HSV-1 contains endogenous RNase activities which are components of the biochemical machinery by which the virus directs host transcription to the synthesis of viral molecules. The uses of virus deficient in the host shut-off function and various drugs were unsuccessful. Previous work in the Millette laboratory demonstrated a sequence in the VP5 promoter that played a significant role in the up regulation of expression of that gene. DNA binding competition studies using a number of HSV-1 sequences exhibiting partial homology to this sequence demonstrated that these sequences all compete for the binding of the same protein factor. Similarly, a piece of the human immunodeficiency virus (HIV) exhibiting a seven base pair homology also exhibited weak competition.

Ph.D. / Molecular Biology / Neurospora crassa arg-2 gene encodes the small subunit of arginine specific carbamoyl phosphate synthethase, the first enzyme in fungal arginine (Arg) biosynthesis. The arg-2 mRNA contains an upstream open reading frame (uORF) specifying an evolutionarily conserved 24-residue peptide called the arginine attenuator peptide (AAP). Synthesis of the AAP causes ribosomes to stall on the mRNA in the presence of high concentrations of Arg. The amino acid sequence of the AAP, and not the sequence of its coding region, is responsible for this regulation. Scanning mutagenesis within the evolutionarily conserved region revealed that some residues are more important than others for the AAP to function. While most known nascent peptides that regulate translation are found encoded as uORFs or as N-terminal leader peptides, the AAP can exert regulatory function whether placed near the N-terminus or internally within a large polypeptide. The AAP’s peptide sensing features are conserved due to regulated stalling of fungal, plant, and animal ribosomes in response to Arg. That the AAP functions as an internal domain to regulate elongation in response to Arg establishes that such domains can provide a means of controlling translational elongation. The minimal sequences required for AAP to function as an internal domain was revealed by systematic deletion of its natural N- and C-terminal regions. Comparative analysis of the AAP with other fungi showed that the evolutionarily conserved region of the peptide is required for regulation. This minimal domain functions when placed as a uORF as seen by toeprint assay. Analyses of Arg analogs provided key insights in the structural requirements for Arg’s role in regulation. These results taken together provide a detailed picture of the requirements for Arg-specific regulation mediated by the AAP.

 As a neuropeptide, both secretin and secretin receptor are expressed in the central nervous system (CNS). It has been revealed that the activities of secretin on hypothalamic cells of rodents are important for osmoregulation and food intake. In the present study, embryonic mouse hypothalamic cell line N42 was used to study the promoter activity of mouse secretin receptor (mSR). By 5′ deletion analysis, a promoter element was identified within ?282 to ?443, relative to the ATG codon, and it contains a GC-box (-297 to -286), a ras responsive element (RRE) (-289 to -276) and an E-box (-416 to -411). Electrophoretic mobility shift assay (EMSA) and supershift analyses showed that Sp1 interacted with the GC-box, another zinc finger As a neuropeptide, both secretin and secretin receptor are expressed in the central nervous system (CNS). It has been revealed that the activities of secretin on hypothalamic cells of rodents are important for osmoregulation and food intake. In the present study, embryonic mouse hypothalamic cell line N42 was used to study the promoter activity of mouse secretin receptor (mSR). By 5′ deletion analysis, a promoter element was identified within ?282 to ?443, relative to the ATG codon, and it contains a GC-box (-297 to -286), a ras responsive element (RRE) (-289 to -276) and an E-box (-416 to -411). Electrophoretic mobility shift assay (EMSA) and supershift analyses showed that Sp1 interacted with the GC-box, another zinc finger / published_or_final_version / Biological Sciences / Doctoral / Doctor of Philosophy

Reconstruction of Transcription Regulatory Network (TRN) and Transcription Factor Activity (TFA) from gene expression data is an important problem in systems biology. Currently, there exist various factor analysis methods for TRN reconstruction, but most approaches have specific assumptions not satisfied by real biological data. Network Component Analysis (NCA) can handle such limitations and is considered to be one of the most effective methods. The prerequisite for NCA is knowledge of the structure of TRN. Such structure can be obtained from ChIP-chip or ChIP-seq experiments, which however have quite limited applications. In order to cope with the difficulty, we resort to heuristic optimization algorithm such as Particle Swarm Optimization (PSO), in order to explore the possible structures of TRN and choose the most plausible one. Regarding the structure estimation problem, we extend classical PSO and propose a novel Probabilistic binary PSO. Furthermore, an improved NCA called FastNCA is adopted to compute the objective function accurately and fast, which enables PSO to run efficiently. Since heuristic optimization cannot guarantee global convergence, we run PSO multiple times and integrate the results. Then GCV-LASSO (Generalized Cross Validation - Least Absolute Shrinkage and Selection Operator) is performed to estimate TRN. We apply our approach and other factor analysis methods on the synthetic data. The results indicate that the proposed PSOFastNCA-GCV-LASSO algorithm gives better estimation. In order to incorporate more prior information on TRN structure and gene expression dynamics in the linear factor analysis model for improved estimation of TRN and TFAs, a linear Bayesian framework is adopted. Under the unified Bayesian framework, Bayesian Linear Sparse Factor Analysis Model (BLSFM) and Bayesian Linear State Space Model (BLSSM) are developed for instantaneous and dynamic TRN, respectively. Various approaches to incorporate partial and ambiguous prior network structure information in the Bayesian framework are proposed to improve performance in practical applications. Furthermore, we propose a novel mechanism for estimating the hyper-parameters of the distribution priors in our BLSFM and BLSSM models, which can significantly improve the estimation compared to traditional ways of hyper-parameter setting. With this development, reasonably good estimation of TFAs and TRN can be obtained even without use of any structure prior of TRN. Extensive numerical experiments are performed to investigate our developed methods under various settings, with comparison to some existing alternative approaches. It is demonstrated that our hyper-parameter estimation method improves the estimation of TFA and TRN in most settings and has superior performance, and that structure priors in general leads to improved estimation performance. Regarding application to real biological data, we execute the PSO-FastNCAGCV-LASSO algorithm developed in the thesis using E. Coli microarray data and obtain sensible estimation of TFAs and TRN. We apply BLSFM without structure priors of TRN, BLSSM without structure priors as well as with partial structure priors to Yeast S. cerevisiae microarray data and obtain a reasonable estimation of TFAs and TRN. / published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

Not available / text

published_or_final_version / Zoology / Master / Master of Philosophy