Parkinson’s disease (PD) is the second most common neurodegenerative disease, affecting 1-2% of individuals above age 65. The mechanisms underlying the pathogenesis of PD are still poorly understood. The most common genetic risk factor for PD is mutations in the gene Glucosidase, Beta Acid 1 (GBA1). GBA1 encodes the lysosomal enzyme glucocerebrosidase, which hydrolyzes glucosylceramide to glucose and ceramide. To study how GBA1 mutations contribute to PD pathogenesis, our laboratory has developed a Drosophila melanogaster model of GBA1 deficiency by introducing a deletion into a Drosophila homolog of human GBA1. My first project utilizes this Drosophila model of GBA1 deficiency to identify genetic modifiers of GBA1 pathogenicity by screening chromosomal deletions for alteration of the GBA1 deficiency phenotype of impaired locomotion and autophagy. Given that only 30% of human GBA1 mutation carriers develop PD by age 80, it is highly likely that GBA1 pathogenicity is modified by genetic factors. For this project, I am currently conducting crosses to obtain flies with the genotypes necessary for the screen. My second project aims to determine whether GBA1-mediated depletion of the lipid ceramide, a biosynthetic precursor to an important class of lipids in membrane composition called sphingosines, contributes to impairment of autophagy and subsequent neurodegeneration through dietary supplementation of ceramide in the Drosophila model of GBA1 deficiency. We hypothesize that ceramide depletion due to low glucocerebrosidase activity contributes to neurodegeneration by lowering cellular sphingosine levels, resulting in altered membrane composition such that fusion of vesicles to the lysosome is impaired. For this project, I conducted one preliminary trial and will conduct further trials aimed at determining an effective method and concentration for dietary supplementation of ceramide in Drosophila. Because lipid metabolism is highly conserved between flies and humans, we expect that genetic modifiers and metabolic effects of GBA1 deficiency identified using the Drosophila model will be applicable to human GBA1-mediated PD.

Eukaryotic cells are constantly locked in genetic conflict with selfish elements, elements that are detrimental to the host, but rely on the host for replication and propagation to progeny. This puts selective pressure on the host to evolve methods of restricting these elements, and selective pressure on the element to evade those methods. Such pressures could cause rapid evolution of host genes, resulting in an evolutionary arms race. One selfish element is the 2-micron plasmid found in budding yeast; in this case, we are referring to Saccharomyces cerevisiae. The 2-micron plasmid relies on the host to propagate, but also confers a 1-3% growth rate defect to the yeast, making it a selfish element and a possible source of genetic conflict. We hypothesize that the 2-micron plasmid may be undergoing an evolutionary arms race with its host, Saccharomyces cerevisiae. To discover if and how an evolutionary arms race is occurring in yeast, we need to look for positive selection in host genes that the plasmid interacts with. Although brewer’s yeast is a very well studied organism, we still do not fully understand how the yeast and plasmid interact. There exists a yeast deletion collection where there is a collection of strains, each with a single non-essential gene deleted. Using the yeast deletion collection, we have discovered some of the host genes the plasmid requires for propagation. This could illuminate possible host evolutionary pathways for evolving restriction mechanisms against a selfish genetic element as well as additional insights about plasmid propagation.

Meier-Gorlin syndrome (MGS) is a rare form of proportionate dwarfism that has been linked to mutations in proteins that function during the initiation and elongation steps of DNA replication. Although mutations in DNA replication proteins have been shown to be causative for MGS, it is still not clear how defects during S-phase lead to the phenotypes observed in patients with this condition. Therefore, further research is necessary to better understand the molecular and cellular consequences of cells harboring MGS mutations. Tools for studying DNA replication in humans are limited, however, there are several well-established tools for studying this process in the budding yeast S. cerevisiae. Some individuals with MGS have been shown to have mutations in the DNA replication initiation factor Cdc45, which functions during both the initiation and elongation steps of DNA replication. Because a specific MGS mutation in CDC45 is in a residue that is conserved between humans and yeast, we have replaced the wild type yeast gene with an allele that contains the equivalent human mutation. Yeast with the MGS mutation exhibit increased sensitivity to elevated temperatures and a reduced ribosomal DNA (rDNA) copy number. In budding yeast, the rDNA locus contains ~150 copies of a 9.1 kb repeat which encodes the template needed to make ribosomal RNA—the main structural component of ribosomes. Additionally, each repeat contains an origin of replication. We hypothesize that this loss of rDNA repeats is a consequence of impaired replication initiation at the rDNA locus and will test this using 2-D gel electrophoresis. We anticipate to expand our study outside of the rDNA locus and examine how replication might be perturbed at other locations across the genome. Ultimately, we hope studying MGS in yeast will elucidate our understanding of the pathogenesis of this condition in humans.

The epithelial-mesenchymal transition (EMT) is a reversible process in which epithelial cells adopt mesenchymal properties by altering their morphology, adhesion, and migratory capacity. During tumor progression, the EMT underlies a process that allows benign tumor cells to infiltrate surrounding tissue and metastasize to distant sites. Our lab has identified a non-canonical Wnt signaling that involves ligands (Wnt 5a/b) and Frizzled2 receptor (Fzd2) in regulating EMT as well as tumor metastasis. Inhibition of Fzd2 signaling suppresses EMT and the progression of metastatic cancer cells both in vitro and in vivo. Recently, we have discovered a non-receptor proline-rich tyrosine kinase 2 (Pyk2) that plays a critical role in Fzd2-mediated cell migration. However, details of Fzd2-mediated activation of Pyk2 and mechanistic understanding of Pyk2 interactions in the pathway still need to be elucidated. We hypothesized that phosphorylation of Fzd2 induces activation of Pyk2, a possible contributor in Fzd2-mediated EMT pathway and tumor metastasis and aimed to uncover 1) the role Pyk2 in Fzd2-mediated EMT induction and cancer metastasis 2) the new mechanistic understanding of how Pyk2 interacts with proteins in the Fzd2 complex. Our preliminary data showed that inhibiting kinase activity of Pyk2 downregulated mesenchymal markers such as vimentin and upregulated epithelial markers such as E-cadherin. Then, the study focused on perturbing Pyk2 expression by RNAi and assessing EMT markers, and cell motility as well as gaining a mechanistic understanding of how Pyk2 interacts with other proteins in the complex. A better understanding of Fzd2-Pyk2 signaling would not only help in connecting critical nodes in Wnt pathway but may open up potential treatment options for metastatic cancers.

DNA stores genetic instructions that are required to program cell activities. Therefore, it is important for DNA to replicate in order to provide a copy of genetic material to daughter cells during mitosis or meiosis. Origin recognition complex (ORC) is a six-subunit assemblage that promotes the initiation of DNA replication. ORC proteins bind in all eukaryotes to a specific sequence found within an origin of replication to assemble and bring together other key initiation factors. Although the ORC proteins themselves are highly conserved throughout evolution, dramatic changes in the origin DNA sequence motif have been observed across species. These changes are most likely accompanied by changes in one or more subunits within the ORC complex. The aim of this project is to understand how changes in the ORC’s binding preferences across species are determined by changes in ORC protein sequences, specifically within the yeast phylogeny. Using genetic engineering approaches, I will introduce foreign ORC alleles into knockout strains of Sacchoromyces cerevisiae which we have generated. Previous data generated by members of our lab and others show how ARS (autonomous replication sequence) Consensus Sequences (ACS, the binding site for ORC) differ across the yeast species. The data indicate that Lachancea kluyveri and Lachancea waltii have intermediate levels of ACS differentiation from Sacchoromyces cerevisiae, so I am introducing alleles from these species as a starting point. Differences in the cell growth rate, size, and fitness will be observed within haploid strains of Sacchoromyces cerevisiae containing the foreign ORC protein in order to determine which ORC subunit(s) is/are responsible for binding preferences for different ACS sequences within the yeast phylogeny. This research can provide a fundamental understanding of evolution of the genome sequence and the proteins that bind it.

 Many genetic mutations have little effect on phenotype. One hypothesis to explain this phenomenon is that strong genetic modifiers buffer genetic and environmental variation. Chromatin remodeling proteins are candidate genetic modifiers because they regulate the transcription of many genes, provide greater robustness to environmental changes, and cause pleiotropic genetic diseases when mutated in humans. To test the hypothesis that chromatin remodeling proteins buffer genetic variation, we will comprehensively assess how deletion of one of three chromatin remodeling proteins (Chd1, Hda1, and Cyc8) alters the phenotypic effect of variants of the transcription factor Ste12. Ste12 is a crucial component of the yeast mating pathway, as it activates the transcription of cell-type-specific genes needed for mating. A mutational scanning experiment of the STE12 gene done in the Fields lab yielded data on the effect on mating of thousands of mutations of STE12. These data provide a baseline for similar scans in strains with a chromatin remodeler deletion. First, we constructed double mutants by deleting each chromatin remodeler gene along with a STE12 deletion. We analyzed the growth rate of each double mutant. The chd1Δ ste12Δ and hda1Δ ste12Δ strains showed healthy growth, but the cyc8Δ ste12Δ strain grew poorly and was flocculent. We then assessed the mating ability of chd1Δ ste12Δ and hda1Δ ste12Δ strains containing the wild type STE12 gene, and found that they could mate sufficiently to continue with the deep mutational scan. We transformed both double mutant strains with a library of plasmids encoding the same Ste12 variants as previously studied, and will calculate a mating score of each variant in the two double deletion backgrounds. If the chromatin remodeler deletion alters the effect on mating of a large number of variants, this result would indicate the importance of chromatin remodeling in phenotypic robustness.

Many inherited diseases are attributed to monogenic mutations; however, some human disorders and conditions are genetically complex, requiring mutations at multiple loci. One of the research goals at the Waterston Lab is to identify genetic interactions that could affect developmental pathways related to human disease. Using the small nematode, Caenorhabditis elegans, as a model organism, we have developed a high-throughput approach to look for interactions. RNAi bacteria were grown separately for knockdown of forty different transcription factor (TF) genes on pooled populations composed of 38 mutated and fully sequenced strains of C. elegans from the Million Mutation Project (MMP). After growth for several generations, the abundance of strains in each condition was compared by targeting strain-specific single nucleotide variants with Molecular Inversion Probes and sequencing on an Illumina MiSeq system. From ~115,000 possible screened genetic interactions, we identified a number of candidate interactions and focused our efforts on a positive interaction between the strain VC20424 and the TF gene B0261.1. While most strains had poor growth on the B0261.1 RNAi, VC20424 had a much higher abundance. We confirmed this candidate interaction by re-testing VC20424 alone and against a control strain on B0261.1 RNAi. We then genetically mapped the VC20424 interaction to a region on chromosome X. In this region we identified a deletion in C25A11.2 which shows similar expression patterns as B0261.1, suggesting a good interaction candidate. We will test additional MMP strains that share the same allele of C25A11.2 on B0261.1 RNAi to look for a similar interaction. We also plan to use CRISPR/Cas-9 to induce the mutation into a control strain and look for conferred resistance to B0261.1, which would further validate our results. Broadly, studying the nature of genetic interactions and potential roles of transcription factors may contribute to the elucidation of human development and complex disease.

Influenza is a prolific and hazardous virus, affecting even those in the population who have been previously infected or vaccinated. The innate immune system serves as a key first line of defense against this pathogen, with the signaling components, called interferons, driving the production of a potent cellular antiviral response. Studies have indicated that viral populations replete in defective virus particles, virions with a deletion in a portion of their genome, are less efficient at blocking the antiviral response, as shown by increased interferon in the host. Our project seeks to explore this phenomenon of RNA deletions leading to increased interferon expression in host cells by testing the hypothesis that deletions in the three polymerase genes of influenza alone are sufficient to cause an increase in the interferon response. To begin answering this question, an interferon reporter system was used to analyze the viral genome of interferon positive cells, and deletions within various lengths of the PB1, PB2, and PA polymerase genes were found to be enriched. I then created pure populations of influenza with these empirically derived gene deletions in PB1 while my mentor, Dr. Alistair Russell, created pure PB2 deletion populations, which were grown on cell lines expressing functional PB1 and PB2 proteins respectively. When used to infect unmodified cell lines, it was found that pure populations of both PB1 and PB2 defective influenza were sufficient to induce the host interferon response. In the future, I will create PA expressing cell lines and influenza with deletions in PA to analyze the effect of PA deletions within the influenza genome on the host interferon response, and it is hoped that these results can be used to explore a mechanism for how influenza occasionally fails to escape the innate immune response.

From blood vessels to the small intestine to the spinal cord, tubes are an essential part of nearly all multi-cellular organisms. Errors in tube formation, called tubulogenesis, cause many of the birth defects that afflict infants, including congenital heart defects and spina bifida, a failure to close the neural tube. Our lab uses the fruit fly Drosophila melanogaster to study tube formation because of the highly conserved nature of this morphological process between our species. My project focuses specifically on a family of genes called Imaginal Disc Growth Factors (IDGFs), which are linked to tubulogenesis in Drosophila and are closely related to a human protein family (CLPs) that have been found to be dysregulated in numerous diseases, including arthritis, and in metastasizing tumor cells. While this homology indicates that IDGFs have a role in cell behaviour, the mechanisms by which these genes act remain unclear. Last year, I used the gene-editing technology CRISPR/Cas9 to investigate the function of one IDGF, the gene Idgf6, by deleting it entirely. Analysis of these knock-out mutants suggests that Idgf6 plays an important role in making and shaping tubes and that removal of Idgf6 results in branched structures instead of discrete tubes. This branching phenotype likely limits the flow of oxygen to the developing embryo, resulting in decreased survival of offspring. My current research explores this genetic pathway further by testing tube morphogenesis under hypoxia and other stress conditions such as crowding, with the aim of further defining the mechanism of tube dysfunction in these mutants.

Successful genome replication is essential for cell division and has been correlated with longevity in the model organism Saccharomyces cerevisiae. DNA replication is initiated at specific sites on chromosomes called origins and requires the recruitment of several low abundance replication factors. The highly repetitive ribosomal DNA (rDNA) locus, known to encode the RNA components of ribosomes in both yeast and humans, has recently been suggested to be influential in genome replication and stability. For instance, the genome of the budding yeast S. cerevisiae contains 450 origins, 150 of which reside at the rDNA locus and could therefore create competition with origins located in the rest of the genome. We hypothesize that the preponderance of rDNA origins compete for essential replication factors and that reduction of rDNA origin competition, achieved by reducing rDNA copy number or compromising the rDNA origin sequence, would promote non-rDNA genome replication. We wanted to examine the role of rDNA competition by using four yeast strains that have either high or low rDNA copy number and either a normal or compromised rDNA origin sequence. We measured the ability of these four strains to maintain a non-essential plasmid containing a non-rDNA origin of replication as an indicator of genome replication. We expect that if rDNA origins compete with other origins, including those on the plasmid, DNA replication will be unable to initiate at the plasmid origin. Therefore, we predict that strains with high rDNA copy number arrays and strong rDNA origins will lose the plasmid at a greater rate than strains with low rDNA copy number and weak rDNA origins. Cells with such intense competition for replication factors could consequently struggle to complete chromosome replication, potentially leading to decreased cellular health and lifespan in yeast, which could be extended to our understanding of human cellular health.

DNA replication is a crucial step in the duplication of all cells, requiring the cooperative function of multiple enzymes. In one of the final steps of chromosomal DNA replication, short pieces of DNA, Okazaki fragments, are joined by the essential enzyme DNA ligase. Mutations in human DNA ligase have been linked to genome instability. To study the consequences of ligase deficiency, my lab uses a temperature-sensitive mutation in the CDC9 gene that encodes DNA ligase in the budding yeast Saccharomyces cerevisiae. When cdc9^ts mutant yeast cells replicate at restrictive temperatures, they lose viability while simultaneously and unexpectedly accumulating loose pieces of single-stranded DNA (ssDNA). The current model of DNA replication cannot explain this accumulation, as newly formed ssDNA fragments should be base-paired to the template strand. We hypothesize that the absence of ligase causes this accumulation of ssDNA, which in turn causes cells to lose viability. I am currently testing this hypothesis in S. cerevisiae cells that have the cdc9^ts mutation as well as a mutation in either EXO1 or RAD27, two genes that code for exonucleases. These exonucleases process Okazaki fragments before ligation, and preliminary evidence suggests that the rad27 mutation decreases viability, while the exo1 mutation improves viability. By growing mutant cells at restrictive temperatures and visualizing ssDNA quantity at multiple time points while determining viability, I am correlating these two phenotypes. If ssDNA causes loss of viability, cdc9^ts rad27 double-mutant cells should have more accumulated ssDNA, correlating to decreased viability, while cdc9^ts exo1 double-mutants should have less accumulated ssDNA. These experiments could provide novel insights into a universal process—DNA replication—and clarify the consequences of defects in that process. Additionally, because ssDNA accumulation has been linked to autoimmune disorders in humans, understanding the causes and consequences of ssDNA accumulation may have implications for human health.

Model yeast Saccharomyces cerevisiae normally grows in the lab as isolated cells but can evolve the ability to clump, protecting the cells from stress factors including antimicrobial treatments. Clumping can occur either due to a failure of cell separation during division or through physical cell adhesion (“flocculation”). Understanding the genetics behind these traits could give us insight into how microbes develop antimicrobial resistance. Previously, we evolved S. cerevisiae from a non-clumping ancestor strain for 300 generations in miniature chemostats, multiplexed culture devices that maintain a constant environment. Using a combination of whole genome sequencing and bulk segregant analysis, a pooling and sequencing approach, we identified causal mutations in 23 independent, clumping clones from the evolution experiments. We identified two strains with cell separation defects which had mutations in both budding gene BEM2 and cell separation gene ACE2. In the remaining clones, the majority of the causal mutations affected the regulation of known flocculation gene FLO1. Despite this shared mutation, a number of these clones exhibited varying flocculation phenotypes. We hypothesized that other “modifier” mutations might also be contributing to the complexity of the phenotype. For one strain with unusual cell morphology, targeted sequencing of candidate modifier genes revealed that the modifier mutation is either in the IRA1 or HSL7 gene, which are genetically linked. The goal of our current experiments is to determine the contributions of BEM2 and ACE2 to the separation defect phenotype and of HSL7 and IRA1 to the flocculation phenotype. We are using a method called complementation testing, in which the wild-type gene is added to the mutated strain using molecular cloning techniques to see if the mutant phenotype is repaired. This will enable us to identify which genes are contributing to the complexity of these traits and facilitate a better understanding of their underlying mechanisms.

In March of 2011 Dr. Bonita Brewer published a paper proposing a model to explain how a particular error in replication can lead to the amplification of chromosomal segments in an inverted orientation, such as those seen in classes of chromosomal rearrangements that arise in human developmental disorders and cancers. She hypothesized that when a replication fork encounters a short inverted repeat, the leading strands could erroneously become ligated to the lagging strand and produce an extrachromosomal palindromic DNA intermediate. Through my research I am investigating if this model can explain the amplification of the ADH4 locus in yeast, what features of the short inverted repeats lead to the inverted junctions, and what environmental or genetic factors influence this mode of gene amplification. Yeast cells that lack a functional copy of ADH1 (a gene that encodes the major form of alcohol dehydrogenase) become sensitive to the drug Antimycin-A. However, these cells can regain Antimycin-A resistance by amplification of ADH4 (a gene that encodes a minor form of alcohol dehydrogenase). Previously it has been found that the extra copies of ADH4 reside on an extrachromosomal palindromic DNA molecule. Through my research I aim to determine if the extrachromosomal amplification of ADH4 is occurring via this new model. To do so, I have created my own adh1∆:KanMX strain, confirmed that high copy number of ADH4 is sufficient to confer resistance to Antimycin-A, and will isolate the extrachromosomal fragment via pulse-field gel electrophoresis and characterize its structure in detail. In an extension of this work I will test whether genetic and environmental factors that stress the replication process influence the frequency of this form of gene amplification, possibly providing insights into amplification events that contribute to human developmental disorders and cancers.

Tyrosine kinases are critical drug targets in oncology due to their role in tumorigenesis. Accordingly, the treatment of specific cancers has been revolutionized by the development of tyrosine kinase inhibitors (TKIs). However, the long-term effectiveness of TKIs is undermined by the emergence of drug resistance. This resistance is often caused by mutations that prevent drug binding. Understanding how a given mutation affects drug sensitivity can enable physicians to personalize their treatment based on a patient's specific tumor genotype. However, current methods to identify resistance mutations are laborious and can only isolate a small subset of possible mutations. Previously, our lab developed a technique to simultaneously measure the functional effect of all single mutations within a protein of interest. We leveraged this method in order to generate a novel system for the rapid and comprehensive characterization of resistance mutations in oncogenic tyrosine kinases, thus overcoming previous technical limitations. Key to this system was choosing a functional assay that could accurately discriminate between the mutations of interest (in our case, drug resistance mutations). Our functional assay was built around Ba/F3 cells; a mammalian cell line whose growth is uniquely dependent on oncogenic kinase signaling. The first step in this project was to validate this growth phenotype by using three clinically relevant oncogenes: FGFR2, ALK, and JAK2. To do this, I cloned each oncogene into a mammalian expression vector, transfected the cloned vectors into Ba/F3 cells, and measured the growth rates of each transfected cell line. As I anticipated, only cells harboring an oncogenic tyrosine kinase grew, thus validating the cell line for further development. These results represent the first step toward the generation of a system for widespread identification of drug resistance mutations in cancer relevant kinases.

Wound healing is a biological process that tissues use to repair themselves after damage. If wound healing is impaired, the tissue may scar or the wound may be fatal. Our lab uses the fruit fly Drosophila melanogaster as a model organism to study the cellular and molecular mechanisms involved in tissue growth, morphogenesis, and homeostasis. Recently while characterizing a novel family of growth factors, Imaginal Disc Growth Factors (IDGFs), our data suggested that one family member, IDGF3, may be upregulated following damage to imaginal discs, the future adult tissue. To explore the potential involvement of the IDGFs in wound healing, I have quantifed their expression by in situ hybridization and localized the mRNA sequences at wound sites in the Drosophila wing imaginal disc. Next, I plan to determine whether specific molecular pathways such as the canonical Wnt pathway or the beta-Akt pathway, which are known to play a role in innate immunity, interact with IDGFs during wound healing. Subsequently, I will use a yeast transcription factor, Gal4, to express RNAi that interferes with mRNA in the same wing disc to determine the relationship between the different pathways. These studies will contribute to our understanding of the mechanisms that regulate tissue healing during development.

Small cell lung tumor (SCLC) is a type of lung tumor that accounts for 10-15% of new lung cancer cases. It is an aggressive cancer that is often not detected before metastasis occurs. Even with chemotherapy and radiotherapy, SCLC has a very high mortality rate and patients tend to develop chemoresistance quickly. Unlike other types of lung cancers, SCLC has no available gene-targeted therapies. This project examines the development of chemoresistance in SCLC genetic mouse models . This approach deletes tumor suppressor genes, such as Rb1 and Tp53, and/or activates oncogenes like lmyc, using a Cre-expressing virus in mouse lung tissue, which leads to the development of SCLC. My role included verifying tumor burden following magnetic resonance imaging (MRI) and quantification of tumor volume in the lungs using MRI data. Once MRI confirmed tumor burden, mice are treated with chemotherapy and tumor volume measured over three weeks of drug treatment. The tumor response to chemotherapy will be compared in different genetic mouse models. At the end of treatment, lung tumors are collected and I use immunohistochemistry to measure the response of the tumors collected from the genetic mouse models. I examine markers of cell proliferation and cell death to assess drug effectiveness. These results could be relevant clinically if certain genetic mutations are found to be associated with chemoresistance.

MSN2 is an environmental stress response transcription factor that is activated in yeast by multiple different stresses—e.g. glucose deprivation, heat, oxidative stress, misfolded proteins among many others—and in turn activates a collection of downstream transcriptional programs. Interestingly, while MSN2 becomes increasingly activated with age, deletion of MSN2 increases the replicative lifespan of the budding yeast. We show that in a glucose-rich environment, MSN2 misreads the environment and drives a pathological glucose starvation response that limits lifespan. Our experiments are facilitated by a novel budget microfluidic/microscopy method rather than the golden standard of microdissection. This innovative system analyzes data through time-lapse images to develop replicative lifespan curves rather than picking off daughter cells with a needle. Microfluidics drastically increases the throughput of experiments and allows for whole-lifespan monitoring of aging yeast cells. The system allows us to efficiently further the research of MSN2 and other transcription factors so that we can potentially translate this to other complex organisms that have similar types of transcription.