Search

Glioblastoma multiforme (GBM) is the most lethal and common type of malignant brain tumor in adults. To date, no curative treatment exists for GBM despite continuous research efforts. Like many other cancers, GBM requires higher levels of methionine for survival compared with normal cells. We aim to exploit GBM methionine dependency as a therapeutic target for this lethal cancer. Our results showed that methionine depletion with an engineered human methionine-γ-lyase (hMGL) reduced GBM cell survival in vitro. Metabolic profiling and MSEA revealed that aminoacyl tRNA biosynthesis, glutathione metabolism, and nucleotide metabolism were significantly changed by hMGL treatment. Mechanistic study showed hMGL treatment resulted in notable increases in oxidative stress in GBM cell lines, leading to DNA damage, and caused cell cycle arrest at the S/G2 phase. In line with this, thioredoxin reductase inhibitor, auranofin, and ATR inhibitor, ceralasertib, showed synergistic effects with hMGL in inhibiting GBM cells. Furthermore, hMGL treatment caused a decrease of global DNA methylation and altered histone methylation patterns. This upregulated the expression of tumor-suppressive microRNAs miR-124 and miR-137, which are frequently silenced in gliomagenesis due to aberrant DNA methylation. Accordingly, hMGL inhibited the phosphorylation and activation of their downstream target, STAT3, a central mediator of GBM growth. Finally, hMGL inhibited the growth of orthotopic human GBM xenografts in vivo and prolonged survival time of tumor-bearing animals. Our data provides strong rationale to investigate the efficacy of hMGL in the treatment for GBM.

T cells are integral to the adaptive immune response against foreign pathogens and tumor cells. They respond to foreign substances, termed antigens, that can elicit immune responses. After becoming activated due to recognition of antigens presented by professional antigen-presenting cells (APCs), T cells can mount multiple effector functions, such as cytotoxic activities, which eliminate infected cells and tumor cells, and helper activities that assist other immune cells’ responsiveness. Memory T cells, which differentiate upon antigen encounter, respond rapidly to subsequent antigen exposure and provide long-lasting protection against pathogens. While one T cell can recognize few antigens, T cells collectively can recognize and respond to a very diverse array of antigens. This unique property by which T cells recognize and respond to a myriad of antigens stems from the rearrangement of their T cell receptor (TCR) genes during development. The highly random TCR rearrangement during development gives each T cell a unique TCR that can recognize a specific antigen, and collectively, the entire T cell repertoire is able to respond to a wide range of antigens. However, autoreactive T cells, which express TCRs that recognizes normal antigens found in the host, could potentially arise from this TCR rearrangement process, posing a potential risk of mounting an autoimmune reaction and damaging the host’s healthy tissues and organs if released to the periphery. The immune system has multiple tolerance mechanisms in place to prevent autoimmunity, and one such tolerance mechanism, termed central tolerance, works by inducing apoptosis of developing autoreactive T cells or by diverting them to the suppressive regulatory T cell (Treg) lineage. T cell development takes place in the thymus. Developing T cells (thymocytes) migrate through the thymic cortex and the medulla at different developmental stages, interacting with and receiving differentiation, proliferation, and selection cues from different thymic stromal cells. After being positively selected for correctly rearranged TCRs that are able to bind self-peptide:MHC complexes expressed by the host, thymocytes migrate from the cortex into the medulla, where they interrogate medullary stromal cells including medullary thymic epithelial cells, dendritic cells (DCs), and B cells. These medullary stromal cells express, process, and present a broad range of self-antigens. Thymocytes that react strongly to the presented self-antigens undergo apoptosis, a process referred to as “negative selection”, or they differentiate into Tregs, thereby removing autoreactive T cells from the naïve T cell repertoire and establishing central tolerance. Chemokine receptor signaling contributes to the orchestrated movement and localization of thymocytes, and their interaction with thymic stromal cells at different development stages. Chemokine receptors are G-protein coupled receptors that respond to a family of small cytokines termed chemokines. Thymocytes upregulate different chemokine receptors at various development stages and receive chemokine cues from different thymic stromal cells. Chemokine receptor signaling plays an important role in thymocytes’ directional movement, speed changes and interactions with thymic stromal cells. Some thymic stromal cells also express chemokine receptors, which may facilitate their own survival, differentiation, and localization within the thymus. The mechanism by which chemokine receptor signaling contributes to the establishment of central tolerance is not fully elucidated. In this dissertation, I will report on the distinct roles of the chemokine receptors CCR4 and CCR7 in the establishment of central tolerance in the thymus. While the role of CCR7 in central tolerance has been relatively well-characterized, the contribution of CCR4 to this process is not well understood. Both CCR4 and CCR7 are expressed by thymocytes after positive selection, raising the question of whether these two chemokine receptors have distinct or redundant impacts on medullary accumulation and tolerance of post-positive selection thymocytes. In Chapter 2, I investigate the roles of CCR4 and CCR7 expression on thymocyte negative selection. Our data reveal a striking disconnect between CCR7 expression and function on intermediate post-positive selection thymocyte subsets, arguing against the consensus in the field that CCR7 expression decisively locates thymocytes to the medulla. Instead, we find that thymocytes express CCR4 within hours of positive selection signaling and swiftly migrate into the thymic medulla in a CCR4-dependent manner, days before CCR7 expression begins. Additionally, CCR4 and CCR7 promote medullary accumulation of immature versus mature post-positive selection thymocyte subsets, and contribute to early versus late phases of negative selection, respectively. The contribution of CCR4 to central tolerance against activated APCs is then highlighted. Collectively, this study suggests a layered model of central tolerance in which CCR4 directs medullary migration and negative selection of immature post-positive selection thymocytes, and CCR7 directs medullary migration and negative selection of more mature post-positive selection thymocytes. The exact mechanism by which chemokine receptors regulate Treg generation in the thymus is not fully understood. In Chapter 3, I investigate the impact of CCR7 expression on thymic dendritic cells (DCs) on the thymic Treg compartment. An increase in Treg generation in thymus was discovered in Ccr7 [superscript -/-] mice. Surprisingly, CCR7 deficiency in the thymic DC compartment, rather than on the thymocytes, contributed to the increased Treg generation in Ccr7 [superscript -/-] mice. CCR7 is expressed by MHCII [superscript hi] thymic conventional DCs (cDCs) of both the Sirpα⁻ cDC1 and Sirpα⁺ cDC2 lineages. Ccr7 deficiency does not alter the cell-intrinsic potential of different DC subsets to induce Treg; however, CCR7 deficiency results in increased apoptosis of MHCII [superscript hi] Sirpα⁻ DCs and a relative increase of MHCII [superscript lo] Sirpα⁺ DCs, which possess the greatest potential to induce Tregs of all thymic DC subsets. This study reveals that CCR7 deficiency alters the composition of the thymic DC compartment, thus dysregulating the induction of Treg. Together, my studies further elucidate the contribution of chemokine receptor signaling to the maintenance of central tolerance in the thymus and provide additional insights into the understanding of mechanisms underlying autoimmune diseases.

The CMP-sialic acid transporter SLC35A1 and UDP-galactose transporter SLC35A2 are two well-characterized nucleotide sugar transporters with distinct substrate specificities. They are in many aspects implicated in human diseases, including congenital disorders of glycosylation. Despite the biomedical relevance, mechanisms of substrate specificity are unclear. To address this critical issue, we utilized a structure-guided mutagenesis strategy and assayed a series of SLC35A1 and SLC35A2 mutants using a rescue approach. Our results suggest that three pockets in the central cavity of each transporter provide substrate specificity. The pockets comprise (1) nucleobase (residues E52, K55, and Y214 of SLC35A1; E75, K78, N235, and G239 of SLC35A2); (2) middle (residues Q101, N102, and T260 of SLC35A1; Q125, N126, Q129, Y130, and Q278 of SLC35A2); and (3) sugar (residues K124, T128, S188, and K272 of SLC35A1; K148, T152, S213, and K297 of SLC35A2) pockets. Within these pockets, two components appear to be especially critical for substrate specificity. Y214 (for SLC35A1) and G239 (for SLC35A2) in the nucleobase pocket appear to discriminate cytosine from uracil. Furthermore, Q129 and Q278 of SLC35A2 in the middle pocket appear to interact specifically with the β-phosphate of UDP while the corresponding A105 and A253 residues in SLC35A1 do not interact with CMP, which lacks a β-phosphate. Overall, our findings contribute to a molecular understanding of substrate specificity and coordination in SLC35A1 and SLC35A2 and have important implications for the understanding and treatment of diseases associated with mutations or dysregulations of these two transporters.

Infectious diseases are emerging at an unprecedent rate in recent years, such as the flu pandemic initialized from Mexico in 2009, the 2014 Ebola epidemic in West Africa, and the 2016-2017 expansion of Zika across Americas. They rarely happened previously and thus lack resources and data to detect and predict their spread. This highlights the challenges in emerging an re-emerging infectious disease surveillance. In the dissertation, I mainly put efforts in developing methods for early detection of such diseases, and assessing predictive power of various models in early phase of an epidemic. In Chapter 2, I developed a two-layer early detection framework which provides early warning of emerging epidemics based on the idea of anomaly detection. The framework could evaluate and identify data sources to achieve the best performance automatically from available data, such as data from the Internet and public health surveillance systems. I demonstrated the framework using historical influenza data in the US, and found that the optimal combination of predictors includes data sources from Google search query and Wikipedia page view. The optimized system is able to detect the onset of seasonal influenza outbreaks an average of 16.4 weeks in advance, and the second wave of the 2009 flu pandemic 5 weeks ahead. In Chapter 3, I extended the framework in Chapter 2 to identify large dengue outbreaks from small ones. The results show that the framework could personalize optimal combinations of predictors for different locations, and an optimal combination for one location might not perform well for other locations. In Chapter 4, I investigated the contribution of different population structures to total epidemic incidence, peak intensity and timing, and also explored the ability of various models with different population structures in predicting epidemic dynamics. The results suggest that heterogeneous contact pattern and direct contacts dominate the evolution of epidemics, and a homogeneous model is not able to provide reliable prediction for an epidemic. In summary, my dissertation not only provides method frameworks for building early detection systems for emerging and re-emerging infectious diseases, but also gives insight to the effects of various models in predicting epidemics.

Plastid genomes of angiosperms are highly conserved in both genome organization and nucleotide substitution rates. Geraniaceae have highly rearranged genomes and elevated nucleotide substitution rates, which provides an attractive system to study nuclear-plastid genome coevolution. My dissertation research has focused on two areas of nuclear-plastid genome coevolution in Geraniaceae. First, I have investigated the correlation of nucleotide substitution rates between nuclear and plastid genes that encode interacting subunits that form the multi-subunit complex of Plastid Encoded RNA Polymerase (PEP). Second, the hypothesis that the unusual changes of plastid genome organization and elevated nucleotide substitution rates of plastid encoded genes is the result of alterations in nuclear encoded DNA replication, recombination and repair (DNA RRR) genes is tested. The second chapter investigates the optimal methods for transcriptome sequencing/assembly. My findings supported the use of transcriptome assemblers optimized for Illumina sequencing platform (Trinity and SOAPdenovo-trans). The third chapter investigated coevolution of nucleotide substitution rates between plastid encoded RNAP (rpoA, rpoB, rpoC1, rpoC2) and nuclear encoded SIG (sig1-6) genes that are part of the multi-subunit complex PEP. Using the transcriptomes of 27 Geraniales species I extracted the PEP genes and performed a systematic correlation test. I detected strong correlations of dN (nonsynonymous substitutions) but not dS (synonymous substitutions) between RNAP and SIG but no correlations were detected for the control genes, which provides a plausible explanation for the cause of plastome-genome incompatibility in Geraniaceae. The fourth chapter investigated the effect of DNA RRR system on the aberrant evolutionary phenomena in Geraniaceae plastid genomes. I extracted DNA RRR and nuclear control genes with different subcellular locations from 27 Geraniales transcriptomes and estimated genome complexity with various measures from plastid genomes of the same species. I detected significant correlations for dN but not dS for three DNA RRR genes, 10 nuclear encoded plastid targeted (NUCP) and three nuclear encoded mitochondrial targeted (NUMT) genes. The findings of a correlation between dN of DNA RRR genes and genome complexity support the hypothesis that changes of plastid genome complexity in Geraniaceae may be caused by dysfunction of DNA RRR systems.

The ability to acquire sufficient iron from the environment is essential for growth of most bacteria, including Escherichia coli and Shigella flexneri. In E. coli, the enterobactin-mediated iron acquisition system is the major way for the cells to get iron under iron-limiting conditions. Enterobactin is a siderophore that is synthesized and secreted in response to iron limitation to scavenge external ferric iron with high affinity. In this work, I showed that the alkyl hydroperoxide reductase (Ahp) system participates in cellular iron metabolism in both E. coli and S. flexneri. The Ahp system is composed of two proteins, AhpC and AhpF. AhpC detoxifies peroxides by converting peroxides to alcohol and water, and AhpF recycles AhpC. In this work, the data showed that the ahpC mutant synthesized and secreted much less enterobactin than the wild type E. coli and had a growth defect in low iron medium. AhpC influenced the first step of enterobactin biosynthesis by either facilitating the delivery of its substrate chorismate to the enterobactin biosynthesis pathway, or maintaining an optimal concentration of chorismate inside E. coli cells. In E. coli, the data showed that deletion of both ahpF and the glutathione reductase gor affected iron uptake or utilization, but not enterobactin biosynthesis, indicating the role of AhpF and Gor in cellular iron metabolism is different from that of AhpC. In S. flexneri, the Ahp system was also found to be involved in cellular iron metabolism; however, AhpC was not required for major steps of S. flexneri virulence: invasion, intracellular replication or cell-cell spread. Overall, the Ahp system participated in multiple steps of cellular iron metabolism.

Drosophila embryonic dorsal-ventral (D-V) polarity is defined by the expression of the gene pipe in the ventral cells of the follicular epithelium surrounding the developing oocyte. pipe encodes a sulfotransferase that transfers sulfate groups to several protein components of the vitelline membrane layer of the eggshell, including Vitelline Membrane-Like protein (VML). These sulfated proteins represent a ventral cue embedded in the eggshell, which, during embryogenesis, leads to the spatially-restricted activation of a serine protease cascade involving Gastrulation Defective (GD), Snake and Easter. Several important pieces of information missing from our understanding of Drosophila D-V patterning include the structures of the carbohydrates borne by the Pipe targets that represent the direct substrates for Pipe and how the Pipe-sulfated ventral cue triggers the spatially-regulated activation of the serine protease cascade. Two major goals of my studies on VML are to elucidate the structures of Pipe-sulfated carbohydrates associated with VML and to identify proteins that interact with VML in a Pipe-dependent manner. To achieve the first goal, I explored different purification systems to isolate VML and eventually found a way of partially purifying VML from Drosophila ovaries. Mass spectrometric analysis of the purified VML is underway to determine the carbohydrate structures on VML and the sites of Pipe-mediated sulfation. Future experiments will involve identification of putative enzymes responsible for the glycosylation of VML and examination of their requirements in D-V patterning. To identify interacting partners of VML, a strategy combining in vivo biotinylation of VML, reversible protein crosslinking and Streptavidin purification of crosslinked complexes will be used. In the second section of my studies, I have demonstrated that VML bearing a biotin acceptor peptide (BAP) tag can be efficiently biotinylated in vivo by co-expressing it with a biotin protein ligase BirA in the follicle cells. As an extension of the application of the approach, I also show that the Torso-like protein, which is localized to the poles of the vitelline membrane and whose polar localization is crucial for Drosophila terminal patterning, can also be biotinylated in vivo when the BAP-tagged protein is co-expressed with BirA in the follicle cells.

It is recognized that an iterative comparative analysis of large-scale homologous RNAs significantly promote the understanding of an RNA family. The Gutell lab is renowned for maintaining high quality RNA sequence alignments and accurately predicted RNA secondary structures using this approach. While the current available alignment and structure data are mainly obtained by trained domain experts with extensive manual effort, it is highly desired that this process is automated and replicable given the exponentially growing number of RNA sequence data and the amount of time required for expert training. In this thesis, we learn the processes involved in comparative analysis by manually aligning a non-coding RNA family, IVS sequences, with the supervision of Dr. Gutell. Each process is then simulated by mathematical objective functions and algorithms. We also evaluate the current available RNA analysis packages that aim each of the processes. Finally, a new RNA sequence alignment algorithm incorporating structure information that can be extended for different alignment tasks is proposed.

In this thesis, a 4.7Å crystal structure of the human mitochondria DNA polymerase γ catalytic complex is reported. Though the DNA substrate-binding site is not identifiable in the structure, two conformational changes in the enzyme architecture are described: 1) rotation of the distal monomer of the accessory subunit towards the catalytic subunit, and 2) shift of the thumb motif of the polymerase domain towards the active site. Both conformational changes suggest a structure of Pol γ in the DNA-bound state and in its active site “closed” conformation.

Protein-protein interactions are central to their biological functions in cells. Many approaches have been applied to study protein-protein interactions in a genomic-scale. In an attempt to develop new strategies to study protein-protein interactions, FRET by using ECFP and EYFP as the donor and receptor was evaluated for possible application in protein-protein interaction study in a high-throughput fashion. Due to the intrinsic properties of ECFP and EYFP, FRET-based protein-protein interaction assay is not suitable for large-scale studies. Instead, tandem affinity purification coupled with mass spectrometry approach proved to be a useful strategy to identify protein interacting partners. Several transcription factor complexes in yeast were successfully purified and novel components in the complexes were identified by combining a shotgun mass spectrometry approach and a differential analysis of the mass spectrometry data. In particular, a negative regulator of G1 to S phase transition during cell cycle, Whi5p, was identified to be a component of SBF complex; a regulator of nitrogen metabolism, Gln3p, was identified to be a component of Hap2/3/5 complex that regulates carbon metabolism, suggesting a crosstalk between nitrogen and carbon metabolism. Additionally, one-step purification coupled with shotgun mass spectrometry analysis was applied to simplify and improve the affinity purification approach used for protein-protein interaction studies. In order to map protein complexes in their native state, a sucrose density gradient was used to separate protein complexes in cells. The proteins within each fraction from the sucrose density gradient were analyzed and quantified with mass spectrometry to obtain the protein abundance profiles across the gradient. The known protein complexes were identified by clustering the protein abundance profiles. This method could possibly be improved to become a generic approach to mapping protein complexes. The goal of protein-protein interaction studies is to determine the protein functions. In an effort to identify ribosome biogenesis genes from a yeast gene network reconstructed from diverse large-scale interaction data sets, at least 25 new ribosome biogenesis genes were confirmed by extensive experimental validations, underscoring the value of proteinprotein interaction studies and gene interaction network.