Chemotherapeutic agents act by targeting rapidly dividing cancer cells. The full extent of their cellular mechanisms, which is essential to balance efficacy and toxicity, is often unclear. In addition, the use of many anticancer drugs is limited by dose-limiting toxicities as well as the development of drug resistance. The work presented in this thesis aims to address the basic biology that underlies these issues through the development and application of chemical genomic tools to probe mechanisms of current and novel anticancer compounds. Chemical genomic screens in the yeast Saccharomyces cerevisiae have been used to successfully identify targets and pathways related to a compound’s mode of action. I applied these screens to examine the mode of action of potential anticancer drugs: a class of platinum-acridine compounds and the apoptosis-inducing compound elesclomol. By analogy to the yeast screens, I developed an RNAi-mediated chemical genomic screen in human cells which has the potential to reveal novel targets and drug mechanisms. This screen was applied to further understand doxorubicin’s mode of action. In parallel with the loss-of-function assays, our lab developed a human ORF overexpression screen in human cells. I applied this gain-of-function screen to identify those genes that, when overexpressed, are toxic to cells. Characterization of such genes that cause toxicity can provide insight into human diseases where gene amplification is prevalent.

Chemotherapeutic agents act by targeting rapidly dividing cancer cells. The full extent of their cellular mechanisms, which is essential to balance efficacy and toxicity, is often unclear. In addition, the use of many anticancer drugs is limited by dose-limiting toxicities as well as the development of drug resistance. The work presented in this thesis aims to address the basic biology that underlies these issues through the development and application of chemical genomic tools to probe mechanisms of current and novel anticancer compounds. Chemical genomic screens in the yeast Saccharomyces cerevisiae have been used to successfully identify targets and pathways related to a compound’s mode of action. I applied these screens to examine the mode of action of potential anticancer drugs: a class of platinum-acridine compounds and the apoptosis-inducing compound elesclomol. By analogy to the yeast screens, I developed an RNAi-mediated chemical genomic screen in human cells which has the potential to reveal novel targets and drug mechanisms. This screen was applied to further understand doxorubicin’s mode of action. In parallel with the loss-of-function assays, our lab developed a human ORF overexpression screen in human cells. I applied this gain-of-function screen to identify those genes that, when overexpressed, are toxic to cells. Characterization of such genes that cause toxicity can provide insight into human diseases where gene amplification is prevalent.

Natural compounds have been largely excluded from characterization via high-throughput profiling strategies due to their limited abundance. Herein, I describe the modification of high-throughput yeast chemical genomic (CG) interaction profiling to permit identifying the modes of action of natural compounds. The previous assay proceeded by evaluating the genome-wide yeast deletion collection for drug-hypersensitivity in a volume of 0.7mL. Compound consumption was minimized with the adapted approach by reducing the assay volume 70% through simplifying the complexity of the yeast deletion pool screened. By recreating each yeast mutant in a drug-hypersensitive background, I created a novel resource that increases compound efficiency and further diminishes compound use. Evaluating a series of characterized compounds analyzed previously by the traditional CG approach validated the adaptations incorporated did not negatively affect the quality of data yielded. Ultimately, this modified strategy will be used to screen thousands of natural compounds contained within the RIKEN NPDepo library.

The past few decades have witnessed an array of advances in DNA science including the introduction of genomics and bioinformatics. The quest for complete genome sequences has driven the development of microarray and massively parallel sequencing technologies at a rapid pace, yielding numerous scientific discoveries. My thesis applies several of these genome-scale technologies to understand genomic response to perturbation as well as chromatin structure, and it is divided into three major studies. The first study describes a method I developed to identify drug targets by overexpressing human genes in yeast. This chemical genomic assay makes use of the human ORFeome collection and oligonucleotide microarrays to identify potential novel human drug targets. My second study applies genome resequencing of yeast that have evolved resistance to antifungal drug combinations. Using massively parallel genomic sequencing, I identified novel genomic variations that were responsible for this resistance and it was confirmed in vivo. Lastly, I report the characterization of chromatin structure in a non-eukaryotic species, an archaeon. The conservation of the nucleosomal landscape in archaea suggests that chromatin is not solely a hallmark of eukaryotes, and that its role in transcriptional regulation is ancient. Together, these 3 studies illustrate how maturation of genomic technology for research applications has great utility for the identification of potential human and antifungal drug targets and offers an encompassing glance at the structure of genomes.

Natural compounds have been largely excluded from characterization via high-throughput profiling strategies due to their limited abundance. Herein, I describe the modification of high-throughput yeast chemical genomic (CG) interaction profiling to permit identifying the modes of action of natural compounds. The previous assay proceeded by evaluating the genome-wide yeast deletion collection for drug-hypersensitivity in a volume of 0.7mL. Compound consumption was minimized with the adapted approach by reducing the assay volume 70% through simplifying the complexity of the yeast deletion pool screened. By recreating each yeast mutant in a drug-hypersensitive background, I created a novel resource that increases compound efficiency and further diminishes compound use. Evaluating a series of characterized compounds analyzed previously by the traditional CG approach validated the adaptations incorporated did not negatively affect the quality of data yielded. Ultimately, this modified strategy will be used to screen thousands of natural compounds contained within the RIKEN NPDepo library.

The past few decades have witnessed an array of advances in DNA science including the introduction of genomics and bioinformatics. The quest for complete genome sequences has driven the development of microarray and massively parallel sequencing technologies at a rapid pace, yielding numerous scientific discoveries. My thesis applies several of these genome-scale technologies to understand genomic response to perturbation as well as chromatin structure, and it is divided into three major studies. The first study describes a method I developed to identify drug targets by overexpressing human genes in yeast. This chemical genomic assay makes use of the human ORFeome collection and oligonucleotide microarrays to identify potential novel human drug targets. My second study applies genome resequencing of yeast that have evolved resistance to antifungal drug combinations. Using massively parallel genomic sequencing, I identified novel genomic variations that were responsible for this resistance and it was confirmed in vivo. Lastly, I report the characterization of chromatin structure in a non-eukaryotic species, an archaeon. The conservation of the nucleosomal landscape in archaea suggests that chromatin is not solely a hallmark of eukaryotes, and that its role in transcriptional regulation is ancient. Together, these 3 studies illustrate how maturation of genomic technology for research applications has great utility for the identification of potential human and antifungal drug targets and offers an encompassing glance at the structure of genomes.

The human gut microbiome is a vast ecosystem of microorganisms that play an important role in human metabolism, immunological function, and even inflammatory gut diseases. Metagenomics research on the human gut microbiome has demonstrated the presence of DNA from dietary yeast species like Saccharomyces cerevisiae. However, it is unknown if the S. cerevisiae detected in metagenomics studies is solely from dead dietary sources or if they can live and colonize the human gut like their close relative Candida albicans. While S. cerevisiae can adapt to low oxygen and acidic environments, it has yet to be explored whether it can metabolize mucin, the primary carbon source found in the mucus layer of the human gut. Mucins are large, gel-forming, highly glycosylated proteins that make up a majority of carbohydrate sources in the gut mucosa. This work determined that S. cerevisiae can utilize mucin as their main carbon source which results in a significant reduction in cell size. Additionally, an aspartyl protease named Yps7, part of a family containing known homologues to mucin-degrading C. albicans proteins in S. cerevisiae, is important for growth on mucin media. To further identify biological pathways required to grow optimally in mucin, both a transcriptome analysis on wild type cells (BY4743) and a chemogenomics screen was performed. In total, 2131 genes demonstrated significant differential expression in mucin media, and 30 genes upon their deletion impacted their growth on mucin. Both these screens suggest that mitochondrial function is required for proper growth in mucin, which was further elucidated by the change in mitochondrial morphology and oxygen consumption in yeast cells upon mucin treatment. Indeed, the uncharacterized open reading frame YCR095W-A is required for growth on mucin as the deletion mutant showed dysfunction in mitochondrial morphology and cellular respiration, further suggesting a potential role in mitochondrial function. Importantly, this project serves as the initial step towards establishing if our most common dietary fungus can survive in the mucus environment of the human gut.

Among the most prevalent and devastating types of human birth defects are those affecting the mouth and face, such as orofacial clefts. Children with malformed orofacial structures undergo multiple surgeries throughout their lifetime and struggle with facial disfigurements, speech, hearing, and eating problems. Therefore, facilitating new research in cranio- and orofacial development is paramount to prevention and treatment of these types of birth defects in humans. Xenopus laevis has emerged as a new tool for dissecting the mechanisms governing facial development. Thus, molecular analyses accompanied by quantitative assessment of morphological changes during orofacial development of this species could be very powerful for understanding how these defects arise. In this dissertation, I present such a study. I first establish a quantitative protocol to describe size and shape changes in facial morphology of wild-type Xenopus embryos. I then utilize this method on embryos in which retinoic acid signaling or folate metabolism have been disrupted to correlate morphological changes with their underlying mechanisms. Finally, I demonstrate the utility of Xenopus as a system for chemical genomics to uncover other regulators of orofacial development.

Upon exposure to heat or cold, Arabidopsis thaliana seedlings undergo rapid transcriptional reprogramming of several hundreds of genes that promote stress tolerance. Despite extensive characterisation of the transcriptional responses to these stimuli, however, relatively little is known about the mechanisms by which temperature signals are perceived and transduced in plant cells. High or low seasonal temperatures have large impacts on crop productivity and are expected to intensify given current global climatic projections. It is therefore of agricultural importance to better understand temperature signalling pathways in plants in order to find solutions to this problem. In this thesis, a chemical genomics screen for molecules activating or repressing heat-inducible genes in A. thaliana was performed in collaboration with Syngenta and the biological targets of these chemicals were predicted based on structural similarities to compounds with known modes of action. Many molecules that affect the function of chloroplasts or mitochondria either activate or repress heat-responsive genes, thus implicating these organelles in the regulation of plant temperature responses. In addition, the translation inhibitor cycloheximide was identified as a repressor of heat-inducible genes and an activator of early cold-inducible genes. Diverse translation inhibitors trigger a cytosolic influx of calcium ions and several inhibitors of translation elongation were found to strongly activate cold-inducible gene expression in a calcium-dependent manner. Furthermore, it was demonstrated that cold shock causes rapid translation repression in A. thaliana seedlings and that the elongation factor LOS1 is involved in cold- or cycloheximide-induced gene expression, thus implicating translational machinery in the regulation of temperature signalling in plants. Finally, one of the chemicals identified in the screen, S01A463859Y, was found to improve heat resilience in A. thaliana and may therefore be of potential use in enhancing crop productivity during thermal stress.

The recently-emerged field of chemical genomics is centered on the use of small molecules to perturb biological systems as a means of investigating their function. In order to employ this approach for the study of plant-pathogen interactions, I established an assay in which Arabidopsis thaliana seedlings are grown in liquid media in 96-well plates. Inoculation of these seedlings with a virulent strain of the bacterial phytopathogen Pseudomonas syringae resulted in macroscopic bleaching of the cotyledons of these seedlings. This symptom was used as the basis for high-throughput chemical genomic screens aimed at identifying small molecules that protect Arabidopsis seedlings from infection. One of the first chemicals identified through this screen was the sulfanilamide compound sulfamethoxazole (Smex). This compound was later shown to also reduce the susceptibility of both Arabidopsis and wheat to infection by the fungal pathogen Fusarium graminearum, suggesting a broad spectrum of activity. More detailed investigations of Smex indicated that the protective activity of this compound did not derive from antimicrobial effects, and that this activity was not executed through common defence-related signalling pathways. The folate biosynthetic pathway enzyme dihydropteroate synthase is a known target of sulfanilamides, and it does appear to contribute to Smex-induced disease resistance, albeit in a folate-independent manner. In order to identify downstream mediators of Smex activity, I initiated two forward genetic screens intended to recover mutants with altered sensitivity to Smex in a seedling growth assay. Interestingly, while these screens yielded mutants with striking Smex sensitivity phenotypes, disease resistance phenotypes were not altered. Gene expression profiling of Arabidopsis tissues treated with Smex prior to bacterial inoculation suggested that this compound generally affects lipid signalling. Altogether, it is evident that Smex elicits a complex set of responses in Arabidopsis with apparently non-overlapping phenotypic outputs.