Browsing by Author "Magwene, Paul M"

Cancer is a major threat to human health and results in 1 in 6 deaths globally. Despite an extraordinary amount of effort and money spent, eradication or control of advanced disease has not yet been achieved. Understanding cancer from an evolutionary point of view may provide new insight to more effective control and treatment of the disease. Cancer as a disease of dynamic, stochastic somatic genomic evolution was first described by Nowell in 1976, and since then researchers have identified clonal expansions and genetic heterogeneity within many different types of neoplasms. The advancement in sequencing technology, especially single-cell sequencing, has open up new frontier by bringing the study of genomes to the cellular level. Phylogenetic analysis, which is a powerful tool inferring evolutionary relationships among various biological species or other entities based upon similarities and differences in their physical or genetic characteristics, has recently been applied to cancer studies and start to show promises in deciphering cancer evolution. However, new challenges have also arisen in experimental design, methodology and interpretation regarding to phylogeny of cancer cells. The overarching theme of this dissertation is to bring phylogenetic analysis to the context of cancer evolution. By using in silico simulations, I show the advantages and disadvantages of different sampling designs for phylogenetic analysis. Although bulk sequencing can hardly recover the topology of phylogenetic trees, I then developed a new method to infer sub-clone spatial distribution utilizing phased haplotypes from bulk sequencing. And lastly, I demonstrate the usage of phylogenetic analysis in breast cancer with multi-regional bulk sequencing and lung cancer with single cell sequencing.

Biological oscillators are ubiquitous in living systems. They allow cellular processes to anticipate and act in synchrony with regular events in the outside world (such as the day/night cycle), or they ensure that processes occur in a particular order. Living things typically contain multiple oscillators, which can often couple to each other and influence each other's timing and function. The purpose of this thesis has been to investigate the relationship between two coupled oscillators in \textit{Saccharomyces cerevisiae}: the yeast metabolic cycle and the cell division cycle. I have focused on two key questions: what is the biological significance of their coupling, and is one oscillator dominant in its interaction with the other?

First, I investigated the temporal relationship between the cell division cycle and metabolic shifts that occur during the metabolic cycle across diverse yeast strains. I showed that a particular cell cycle event (DNA replication) was consistently delayed relative to a metabolic event (entry into the high oxygen consumption phase). This suggested that an earlier cell cycle event (Start and commitment to the cell cycle) was tied to the onset of high oxygen consumption. Second, I used fluorescent probes to examine the relationship between the metabolic cycle and the commitment to cell cycle progression at single-cell resolution. This revealed that cells enter high oxygen consumption phase of the metabolic cycle before passing Start, supporting a model of metabolic cycle/cell division cycle coupling in which the shorter metabolic cycle controls cell cycle commitment, likely via modulation of cell size thresholds.

Evolutionary dynamics can be notoriously complex and difficult to analyze. In this dissertation I describe a population genetic regime where the dynamics are simple enough to allow a relatively complete and elegant treatment. Consider a haploid, asexual population, where each possible genotype has been assigned a fitness. When mutations enter a population sufficiently rarely, we can model the evolution of this population as a Markov chain where the population jumps from one genotype to another at the birth of each new mutant destined for fixation. Furthermore, if the mutation rates are assigned in such a manner that the Markov chain is reversible when all genotypes are assigned the same fitness, then it is still reversible when genotypes are assigned differing fitnesses.

The key insight is that this Markov chain can be analyzed using the spectral theory of finite-state, reversible Markov chains. I describe the spectral decomposition of the transition matrix and use it to build a general framework with which I address a variety of both classical and novel topics. These topics include a method for creating low-dimensional visualizations of fitness landscapes; a measure of how easy it is for the evolutionary process to `find' a specific genotype or phenotype; the index of dispersion of the molecular clock and its generalizations; a definition for the neighborhood of a genotype based on evolutionary dynamics; and the expected fitness and number of substitutions that have occurred given that a population has been evolving on the fitness landscape for a given period of time. I apply these various analyses to both a simple one-codon fitness landscape and to a large neutral network derived from computational RNA secondary structure predictions.

Bet-hedging strategies are used by organisms to survive in

unpredictable environments. To pursue a bet-hedging strategy, an

organism must produce multiple phenotypes from a single genotype. What

molecular mechanisms allow this to happen? To address this question, I

created a synthetic system that displays bet-hedging behavior, and

developed a new technique called `TrackScar' to measure the fitness

and stress-resistance of individual cells. I found that bet-hedging

can be generated by actively sensing the environment, and that

bet-hedging strategies based on active sensing need not be

metabolically costly. These results suggest that to understand how

bet-hedging strategies are produced, microorganisms must be

examined in the actual environments that they come from.

In an attempt to gain more insight on functional evolution of cAMP-PKA pathway I have taken a comparative approach and examined functional interactions of cAMP-PKA signaling in well-studied yeast developmental programs and closely related Saccharomyces sensu stricto. species. I have shown that variation in cAMP-PKA signaling contributes significantly to variation in developmental responses in S cerevisiae. Variation in pseudohyphal growth and sporulation, two inversely correlated developmental strategies to nutrient limitation in yeast, proportional to variation in intracellular cAMP levels. S. cerevisiae strains proficient in pseudohyphal growth have higher intracellular cAMP concentrations relative to strains that sporulate efficiently. Phenotypic, genetic and signaling data presented here suggest that the cAMP-PKA signaling underlies a phenotypic trade-off between sporulation and pseudohyphal growth in S. cerevisiae.

Further investigation into the role of cAMP-PKA signaling in closely related S paradoxus and S bayanus revealed an antagonistic function of cAMP-PKA signaling for developmental responses in S. bayanus. Unlike in S. cerevisiae, increased cAMP concentrations surprisingly inhibit pseudohyphal response in S. bayanus. Another unanticipated finding in this work is that in S. bayanus. Flo11, required for pseudohyphal differentiation in S. cerevisiae, is dispensable. Additionally, interactions of cAMP-PKA signaling and the general-stress response mechanism appear reversed in S. bayanus. As shown by deletion mutation, gene expression and pharmacological treatment data, altered interactions and alternative targets downstream of cAMP-PKA could critically contribute to alternative regulation of nutrient-induced development in S. bayanus.

Intracellular cAMP concentrations show decaying oscillations upon glucose replenishment in derepressed yeast cells. The quantitative characteristics of oscillations are distinct within and between Saccharomyces species. Given the tight regulation of cAMP levels and its critical role, the variation in cAMP oscillatory dynamics could be reflective of differential interactions of cAMP-PKA signaling that also underlie induction of developmental programs to changing environments. As such, intracellular cAMP levels and dynamics could potentially be used as molecular phenotypes.

Seemingly against all odds, some of the microscopic larvae of deep-sea chemosymbiotic fauna make it through the largest ecosystem on earth to settle at a new population. Evidence of these migration events can be found in the distribution of extant populations, by capturing larvae in the water column and by examining the genetic makeup of adults. In this dissertation I use large genetic datasets to consider how migration and gene flow have influenced the population structure of two deep-sea mollusks. Specifically, I assess if observed connectivity is the result of migration mediated by stepping-stones, recent divergence and/or long distance dispersal.

In Chapters 1 and 4, I examine gene flow in vesicomyid clams which live on reducing habitats like hydrothermal vents, cold seeps and food falls around the world. Of particular interest is the clam Abyssogena southwardae. This clam has colonized habitats across the entire Atlantic Ocean basin and exhibit low genetic diversity between populations. I test the possibility that the Mid-Atlantic Ridge has served as a stepping-stone across the basin using cytochrome c oxidase subunit I mitochondrial sequences (Chapter 1) and 5 additional nuclear loci (Chapter 2). The results of this analysis suggest that the Mid-Atlantic Ridge does not serve as a stepping-stone. An alternative hypothesis for the low genetic diversity in the Abyssogena southwardae is recent colonization of these populations by a common source.

In Chapter 2 I ask a similar question about the connectivity between distant populations of the hydrothermal vent endemic limpet Lepetodrilus aff. schrolli. Despite being separated by thousands of kilometers and exhibiting strong population structure, the two populations of Lepetodrilus aff. schrolli show evidence of directional gene flow since population divergence. The lack of identifiable migrants and the modeled ocean currents in the region suggest that it is unlikely that a single larva could make the trip between the two basins. Instead, one or more of the active hydrothermal vents in the vicinity could serve as a stepping-stone between the two distant basins. I was able to detect directional migration in this non-equilibrium scenario due a large dataset consisting of 42 amplified gene regions across 93 individuals.

In Chapter 3 I explore how subsets of this large amplicon dataset can provide insight into how the number of loci sequenced affects accuracy, precision, type I error and type II error. This subset analysis shows a general increase in accuracy and precision and a general decrease in type I and type II error with the addition of more loci. Furthermore, in the particular case of Lepetodrilus aff. schrolli subsamples as large as 20 loci still gave inaccurate but precise estimates. By examining the gene trees of these amplicons I was also able to investigate how empirical gene tree discordance influences coalescent analysis.

Large amplicon datasets allowed me to detect gene flow in complicated demographic situations in Abyssogena southwardae and Lepetodrilus aff. schrolli. In the Atlantic, Abyssogena southwardae analysis of multiple loci suggest that there are not stepping-stones along the Mid-Atlantic ridge to connect the Atlantic Equatorial Belt. This analysis will benefit from the additional loci that have been collected but not analyzed. In the Pacific, Lepetodrilus aff. schrolli, stepping-stones seem like a much more likely hypothesis to accommodate the low level of migration seen between the two distant basins examined. These studies are among the largest genetic datasets collected to date in the deep-sea and should serve as a benchmark for the generation and analysis of datasets for studying connectivity in the deep-sea.

Fungi of the basidiomycete genus Cryptococcus cause disease in an estimated quarter of a million people, annually. Cryptococcus neoformans and Cryptococcus deneoformans are the two most prevalent disease causing species within the Cryptococcus clade, with isolates of these species exhibiting considerable variation in their pathogenicity, ranging from benign to highly virulent. A wide variety of traits, such as thermal tolerance, melanin production, and an extracellular capsule contribute to virulence, yet our understanding of the genetic architecture of such traits is limited. In the studies reported here, I describe the first genome-wide analyses of recombination in C. neoformans and C. deneoformans and provide the first high-resolution genetic mapping studies of virulence traits in these important fungal pathogens.

In studying recombination, I considered both the nuclear and mitochondrial genomes, and estimated recombination rates for both opposite- and same-sex matings. With respect to recombination of the nuclear genome, I found that progeny from opposite-sex mating have more crossovers on average than those from same-sex mating. These analyses also suggest differences in recombination rate between C. neoformans and C. deneoformans. Similarly, analyses of mitochondrial inheritance and recombination point to differences between offspring from opposite- and same-sex matings, though with much lower overall rates of recombination as compared to the nuclear genome.

To dissect the genetic architecture of complex virulence traits, I employed quantitative trait locus (QTL) mapping. A unique aspect of these QTL studies was the application of functional data analysis methods that exploit time-series data and multiple experimental conditions. I mapped QTL for thermal tolerance, melanization, capsule size, salt tolerance, and antifungal drug susceptibility in C. deneoformans. For several QTL, I was able to identify candidate causal variants that underlie these loci. Two major effect QTL for amphotericin B resistance map to SSK1 and SSK2; regulators of the high osmolarity glycerol (HOG) pathway that governs responses to osmotic stress. Epistatic interactions between SSK1 and SSK2 were also shown to govern fludioxonil sensitivity. A third major effect, pleiotropic QTL was mapped to the gene, RIC8, a regulator of cAMP-PKA signaling. RIC8 variation is predicted to contribute to differences in thermal tolerance, melanin production, and capsule size.

In combination, the studies reported here advance our understanding of the mechanisms that generate and maintain variation in Cryptococcus and implicate genetic variants in key stress-responsive signaling pathways as a major contributor to phenotypic variation between lineages of Cryptococcus.

Adapting to environmental changes (i.e. an increase in osmolarity) is critical for cell survival. How cells respond and adapt to osmotic stress has been well-studied in the model eukaryote Saccharomyces cerevisiae. Although the molecular and systems properties of osmo-adaptation have been well studied, few studies have focused on the effects of genetic variation. Understanding how genetic variation affects molecular pathways and their dynamics, which translates to variation in cellular and organismal phenotypes, is a key step towards understanding important phenomena such as complex gene by gene interactions and the mapping of genotype to phenotype. The challenge is to causally connect genetic differences with cellular function and differences in complex traits between individuals. As a first step towards addressing this challenge, my dissertation research investigates how natural genetic variation affects osmo-adaptation dynamics in budding yeast, Saccharomyces cerevisiae.

First, I characterized the natural variation in osmo-adaptation dynamics across S. cerevisiae. I showed that individual strains were highly variable in adaptation time and relative maximum growth rate after adapting to stress. Analysis of a broad set of genes involved in osmo-adaptation did not reveal any obvious genetic differences that could account for the observed variation. To identify alleles associated with the variation, I measured osmo-adaptation dynamics in progeny generated from a cross between two closely related lab strains. Identified alleles were outside the core signaling pathway and affected both adaptation time and relative maximum growth rate. Finally, I built a novel mapping panel and measured osmo-adaptation dynamics to obtain a more global, species-wide view. The panel showed an increased amount of variation in osmo-adaptation dynamics and a subset of progeny were phenotypically more extreme than their parents. Mapping the variation in this panel will generate a comprehensive list of alleles that affect osmo-adaptation. The strains in the mapping panel have a low number of mutations predicted to have strong effects in HOG pathway genes. Given our earlier results from the pairwise cross, I expect that many osmo-adaptation alleles discovered from the mapping panel will be outside the HOG pathway.

The evolution of pathogenesis, in many cases, is a story of competition between host and microbe; however, many opportunistic pathogens are primarily found in niches other than the host environment. Such pathogens frequently lack host-to-host transmission, and there may be limited opportunities for an infectious population to be re-dispersed back into the environment. Observations such as these motivate the hypothesis that the evolution of virulence traits in opportunistic pathogens may be primarily driven by environmental selective pressures, rather than the host-environment per se.

For Cryptococcus the ability to survive interactions with macrophages and the ability to grow at host body temperatures are indispensable to its pathogenic capabilities. The work presented here aims to dissect the genetic underpinnings of these virulence traits using the abundant natural variation of Cryptococcus and using the accumulation de novo mutations associated with growth under relevant stressors.

An important aspect of the hypotheses surrounding Cryptococcus evolution is the predator-prey interactions it has with free-living amoeba. Amoebae are able to consume Cryptococcus cells in a manner similar to how macrophages phagocytose and digest infectious cells. This similarity is the basis of the “Amoeboid Predator-Fungal Animal Virulence Hypothesis” which posits that amoeba act as training grounds for environmental fungal pathogens and thus inadvertently select for resistance to immune phagocytes. I tested this hypothesis by using QTL mapping to identify genes and alleles that are involved in amoebae resistance in both C. neoformans and C. deneoformans. I identified QTL that contribute to amoeba resistance, and discovered that the largest effect QTL in both species localize to homologous regions of the genome, suggesting a shared mechanism of amoeba resistance. In C. neoformans, this QTL also contributes to variation in melanization. I identified a causal variant for this QTL, a non-coding deletion upstream of a transcription factor, BZP4. Contrary to the predictions of the Amoeboid Predator-Fungal Animal Virulence Hypothesis, I did not find an association between the ability to survive amoeba predation and virulence in either in vitro or in vivo models of infection. These findings suggest a re-evaluation of the amoeba predation model for the evolution of pathogenesis, suggesting that factors other than amoeba may provide the significant selective pressures that underlie virulence ability.

I extended my quantitative analyses of Cryptococcus to two important factors involved in both environmental and disease contexts: thermal and low pH tolerance. In doing so, I discovered multiple pleiotropic QTL involved in general growth that also dictate stress tolerance in both high temperature and low pH environments. By fitting growth data to a Gompertz growth model and QTL mapping based on the parameters of this model, I discovered a novel QTL that effects lag, the time it takes for a population of cells to begin growing at an exponential rate. This lag QTL is pleiotropic across growth conditions. I identified a candidate allele for the lag QTL, a 9-bp deletion in CNAG_01111, a gene that has been found to impact growth initiation in other species of fungi.

Finally, taking a complimentary approach to understanding the role of genes in environmental survival, I experimentally evolved a C. neoformans strain in conditions of thermal stress and fludioxonil stress. I discovered that strains evolved at high temperatures lose tolerance to fludioxonil and strains evolved in fludioxonil lose temperature tolerance. Furthermore, the loss of fludioxonil tolerance in the high temperature evolved strains can be partially rescued by growing them on media containing fludioxonil. This rescue results in a proportional loss of thermal tolerance. Studying the genomic changes behind the evolved phenotypes I discovered multiple large scale deletions and one multi-gene duplication associated with fludioxonil resistance and a single multi-gene deletion associated with thermal tolerance. There are also a variety of small scale mutations associated with each evolved condition, including mutations of genes in the HOG and ergosterol pathway that are responsible for fludioxonil resistance. Mutations in uncharacterized multidrug transporters are frequently associated with fludioxonil resistance, suggesting that the evolved strains might also have altered resistance to other antifungals. These findings highlight the polygenic and pleiotropic genetic architecture of adaptation in C. neoformans on an ever warming planet with increased use of agricultural antifungals. The trade-offs found may represent a good sign for the use of phenylpyrroles as an agricultural antifungal.

Collectively, my work sheds light on genes and alleles involved in environmental survival while also making important connections back to human disease. It also exhibits the importance of utilizing the natural variation of fungal pathogens to study the evolutionary hypothesis surrounding virulence traits. The studies reported here also provide significant groundwork for many new insights into virulence genes and the origins of Cryptococcus pathogenicity.

Humanity is shaped by its relationships with microbes. From bacterial infections to the production of biofuels, industry and health often hinge on our control of microbial populations. Understanding the physiological and genetic basis of their behaviors is therefore of the highest importance. To this end I have investigated the genetic basis of plastic adhesion in Saccharomyces cerevisiae, the mechanistic and evolutionary dynamics of mixed species biofilms with Escherichia coli and S. cerevisiae, and the induction of filamentation in E. coli. Using a bulk segregant analysis on experimentally evolved populations, I detected 28 genes that are likely to mediate plastic adhesion in S. cerevisiae. With a variety of imaging and culture manipulation techniques, I found that particular strains of E. coli are capable of inducing flocculation and macroscopic biofilm formation via coaggregation with yeast. I also employed experimental evolution and microbial demography techniques to find that selection for mixed species biofilm association leads to lower fecundity in S. cerevisiae. Using culture manipulation and imaging techniques, I also found that E. coli are capable of inducing a filamentous phenotype with a secreted signal that has many of the qualities of a quorum sensing molecule.