Browsing by Author "Donohue, Kathleen"

As habitat fragmentation and changing climatic conditions continues to pose threats to species persistence, it is important to study the traits that may ensure survival of individual plants and the population-level effects of those traits. Previous studies have highlighted the importance of one such trait, dispersal, for population persistence under changing conditions. Dispersal allows plants to track spatial and seasonal changes in the environment. Plants can disperse through space, i.e., pollen and seed dispersal, and through time, i.e., delaying germination within seasons (seed dormancy) or between years (seed banking). Spatial sorting of good dispersers has been highlighted as a mechanism to facilitate spatial habitat tracking, but few studies have evaluated how interactions between spatial sorting and plasticity of dispersal-related traits may interact to impact dispersal dynamics. In addition, theory predicts that both spatial dispersal as a form of bet-hedging, and seasonal seed dormancy as a form of habitat selection, may stabilize population demography across years, thereby reducing population extinction risk. Few studies have experimentally manipulated spatial and temporal dispersal in the field to test these theoretical predictions. My dissertation quantifies spatial dispersal at a small scale, tests for the genetic basis and plasticity of dispersal-related traits, tests the metapopulation consequences of local spatial dispersal versus spatial isolation, and quantifies the population-level consequences of seasonal seed dormancy. I combined field, greenhouse, and quantitative genetic approaches to assess the demographic effects of spatial and temporal dispersal in the model plant species Arabidopsis thaliana. In my first chapter I show that limited dispersal results in predictable variation in post-dispersal density across space. Further, I show that the traits that enhance dispersal ability in the field have a genetic basis, exhibit plasticity to density and season length, and genetic variation in that plasticity. Plasticity and genotypic differences in plasticity can alter the effects of spatial sorting on dispersal ability across a species range. Specifically, plasticity can augment the effects of spatial sorting of genotypes, by enhancing dispersal at low-post-dispersal density. Genetic differences in plasticity of good versus poor dispersers can mask genetic differences in dispersal ability and thereby slow spatial sorting of genotypes at high density but augment genetic differences and spatial sorting at low density. In my second chapter I show that, compared to isolated populations, populations open to dispersal had smaller between-year population size fluctuations, increased survival of individuals within years, and less between-population differentiation in morphological traits. Those demographic effects of dispersal may have increased the effective population size of populations open to dispersal, facilitating a recovery from the effects of harsh environmental conditions. Finally, my third chapter demonstrates that seasonal seed dormancy may allow populations to better take advantage of favorable conditions by increasing population size and stabilizing population demography over time in more permissive environments. However, contrary to expectations, dormancy did not reduce the effects of environmental variation due to its inability to counter the effects of population bottlenecks induced by harsh environmental conditions. The findings of my dissertation highlight the importance of considering the interaction between spatial sorting, phenotypic plasticity, and genetic variation in plasticity when projecting range expansion dynamics. My dissertation also provides some of the first experimental validation of the theory that predicts that spatial and temporal dispersal can stabilize population demography and facilitate population persistence. Therefore, when predicting how species may respond to anthropogenic changes it is important to not only consider the effects of dispersal on environmental tracking, but also the effects of dispersal on population demography. Finally, given the limited dispersal of Arabidopsis thaliana and many annual species like it, it is important to measure population dynamics at a micro-scale, otherwise researchers risk underestimating regional-level extinction risk.

Environmentally cued development is widespread across the natural world. Many organisms rely on abiotic and biotic cues to undergo developmental transitions like budburst, flowering, and mating at the appropriate times of year. The study of the timing of these transitions is known as phenology. Because the timing of development determines the environment experienced by the next life history stage, it has the potential to affect evolutionary processes that occur after development. Further, because the timing of development can filter the environment experienced by the next life history stage, it can be considered a form of “habitat tracking.” In this dissertation, I use manipulative laboratory and field experiments to quantify how germination phenology can alter the postgermination environment, show that the postgermination environment can itself be genetically determined, show that germination phenology is a form of habitat tracking, and test how germination phenology can affect trait expression, natural selection, and fitness.

In my first chapter, using the ecological and genetic model species Arabidopsis thaliana, I showed that when the timing of development is genetically controlled, and the timing of development affects the environment experienced by the next developmental stage, then the environment experienced after development can itself be inherited and can evolve. Further, I demonstrated that germination phenology is a form of “habitat tracking”, by enabling seeds to establish seedlings in a subset of the full environmental conditions available. In my second chapter, using ecologically diverse A. thaliana genotypes, I show that the timing of germination can affect natural selection on postgermination traits, and that postgermination traits can affect selection on germination phenology. In my third chapter, using two plants native to North Carolina, Phacelia purshii and P. fimbriata, I show that populations can vary naturally in their propensity to germinate in response to different environmental cues, that populations preferentially germinate in habitats that are beneficial for seedlings, and when placed in new geographic locations, populations may use phenology to track novel but beneficial environmental conditions.

My dissertation placed the common process of cued development into the well- established theoretical framework of habitat tracking and habitat selection. By doing so, I was able to generate and test novel predictions about potential consequences of phenological cueing that have not yet been explored—namely, that the post- development environment itself can be inherited, that the magnitude and frequency of natural selection can vary with changes in habitat tracking, that habitat tracking itself may evolve in response to traits expressed and environments experienced after development, and that habitat tracking through phenology may be an important mechanism that organisms can use to cope with climate change.

While phenotypic plasticity has been widely documented, the relative contribution of parent versus offspring environment to determining progeny phenotypes and the persistence of parental effects throughout progeny development under different environments remain unknown. In predictable and/or seasonal environments, parental effects are predicted to be favored during early life stages of offspring, while offspring environment is predicted to increase in relative importance as the accuracy of offspring perception increases over the course of development. Furthermore, as plastic responses to the environment are both transmissible across generations and dynamic across development, epigenetic mechanisms are likely involved in the regulation of parental effects and phenotypic plasticity. The neighboring community is an important environmental factor for plants since sessile plants cannot escape competition within their current generation. Whether interactions with neighbors are positive or negative likely depends on the environment, as competition is hypothesized to increase in favorable environments and the neighboring community of plants may change over a seasonal progression. For these reasons, neighbor environment is an interesting and ecologically important environmental variable that can be used to investigate how parental and progeny environment regulate progeny phenotypes throughout development. To test these predictions, this dissertation evaluated how progeny phenotypes responded to the combination of parental and offspring environments, quantified how parental and progeny methylation regulate offspring phenotypes, and examined their effects on plasticity. To address which generation’s environment and DNA methylation affect phenotypes in offspring, I manipulated simulated and real neighbor environments and DNA methylation within and across generations in Arabidopsis thaliana, a winter annual native to Eurasia widely introduced across North America. In Chapter 1, using chemical demethylation, I tested whether parental and progeny DNA methylation influences germination, and whether parental DNA regulates germination response to past and present simulated canopy. I found that germination of offspring is regulated by parental DNA methylation and is responsive to parent, not seed, environment for most genotypes. Furthermore, I confirmed using mutant lines that all contexts of DNA methylation were involved in the transmission of parental effects, but they may operate through different pathways controlling germination. In Chapter 2, I quantified how parental versus progeny methylation regulate progeny phenotypic responses to parental and progeny canopy shade. I found that both parent and offspring canopy affect offspring traits across development, but parental environment had stronger effects at the seedling stage. Both parent and offspring DNA methylation affected offspring response to canopy, but parental DNA methylation only affected traits at the seedling stage. Trait correlations were significantly altered by chemical demethylation of parents and offspring, indicating that DNA methylation of both generations are important in regulating development and integrating phenotypic response to canopy. Finally, in Chapter 3, I tested whether parent or offspring DNA methylation affected response to the heterospecific neighbor, Stellaria media, under simulated seasonal conditions. I found that growing with competitors decreased fitness for all genotypes, but genotypes varied in the effect of neighbors on morphology and fitness. Both parent and offspring DNA methylation had direct effects on growth and fitness in all genotypes, but genotypes varied in how DNA methylation influenced response to neighbors. In one genotype, plastic response to neighbors was unaffected by chemical demethylation treatments, indicating that neighbor-induced plasticity is not always mediated via DNA methylation. Together, these results indicated that offspring phenotypes are shaped by both parent and offspring environment, and that parental environment and parental DNA methylation are especially important in regulating offspring traits early in life. The genetic variation observed in the expression of phenotypic plasticity via parental and progeny DNA methylation suggests that the epigenetic regulation of progeny phenotypes has a genetic basis and may evolve.

Theoretical and empirical studies have consistently shown that the optimal timing of seed germination reduces exposure to physical stress and minimizes competitive interactions with neighbors. However, this research has not accounted for facilitative (positive) interactions among plants, which become more pronounced as environmental stress increases. Facilitation is more likely to occur early in a plant's life when it is more susceptible to stress. In seasonal environments, the stress a given individual experiences can change throughout the year, and some years are more stressful than others. These sources of temporal variation in stress will dictate the facilitation-competition balance that individuals experience. However, it remains unclear how this balance affects the optimal timing of germination. My dissertation research asks how the timing of germination responds to neighbors, how those responses affect the facilitation-competition balance individuals experience, and how that balance in turn affects fitness and demography. More generally, it asks how the timing of germination and other types of emergence affect the facilitation and competition that individuals experience throughout their lives.

I used laboratory, greenhouse, and field experiments to examine how the timing of germination in the winter annual Arabidopsis thaliana (Brassicaceae) responds to cues of neighbors and how those responses affect interactions with neighbors. I then developed a mathematical model of population growth in an annual plant to examine how intraspecific facilitation and competition over ontogeny affect the optimal degree of investment in dormancy (i.e., delayed germination) in variable environments.

My experiments revealed that seeds of A. thaliana typically delay germination in response to neighbors and that these responses can promote facilitative interactions and reduce competitive ones with neighbors. Selection against delayed germination, which occurs because of stress later in the season, can be mitigated by facilitation. Further, delaying germination can be beneficial by increasing the difference in sizes between seedlings and their neighbors, which may promote resource partitioning. In the theoretical study, I found that increasing the degree of investment in the fraction of dormant seeds (i.e., delaying germination) can promote the persistence of populations that experience both facilitation and competition in variable environments. This occurs because increased dormancy prevents high juvenile densities that promote facilitation and consequently limit reproduction in large populations. The findings of this research indicate that plant-plant interactions depend strongly on temporal context, and they reveal that the facilitation-competition balance determined by temporal variation in stress plays a key role in how germination and dormancy traits will evolve in variable environments.

Angiosperms have evolved multiple mating systems that allow reproductive success under varied conditions. Striking among these are cleistogamous mating systems, where individuals can produce alternative flower types specialized for distinct mating strategies. Cleistogamy is thought to be environmentally-dependent, but little is known about environmental triggers that induce cleistogamous flower or the gene regulatory networks that determine the final floral phenotypes. If production of alternate flowers is environmentally induced, populations may evolve locally adapted responses. Mimulus douglasii, exhibits a cleistogamous mating system, and ranges across temperature and day length gradients, providing an ideal system to investigate environmental parameters that control cleistogamy and the gene regulatory networks responsible for the different floral forms. In these studies, we compared flowering responses across M. douglasii population accessions that produce phenotypically distinct outcrossing, and self-pollinating flower morphs. Under controlled conditions, we determined time to flower, and number and type of flowers produced under different temperatures and day lengths. We also compared gene expression profiles between chasmogamous and cleistogamous flowers using RNA-seq. We find that temperature and day length both effect onset of flowering. Long days shift flower type from predominantly chasmogamous to cleistogamous. The strength of the response to day length varies across accessions whether temperature varies or is held constant. We also find that gene expression patterns differ between the early development chasmogamous and cleistogamous flower buds. Cleistogamy is an environmentally sensitive polyphenism in Mimulus douglasii, allowing transition from one mating strategy to another. Longer days induce flowering and production of cleistogamous flowers. Shorter days induce chasmogamous flowers. Population origin has a small effect on response to environmental cues. Subtle shifts in the expression of cell division, cell expansion, and metabolic process related transcripts lead to the massive size difference observed between chasmogamous and cleistogamous flowers.

This dissertation examines the processes that generate phenotypic variation in life cycles in seasonal environments. Collectively, a life cycle describes the stages an organism passes through during a generation. The timing, or phenology, of these transitions is often influenced by both environmental and allelic variation. Using the model organism Arabidopsis thaliana and both empirical and modeling approaches, I examine how correlations between life-cycle transitions, environment-dependent allelic effects, and epistasis generate patterns of life-cycle variation both within and between generations. In my first chapter, I use experiments to determine that many combinations of genetic, environmental, and developmental factors can create similar germination phenotypes, that maternal effects can influence phenotypes more than genetic differences, and that cross-generational effects can reduce variation in germination timing despite variation in flowering and dispersal time. In my second chapter, I use a modeling approach to consider the entire life cycle. I find that environmental variation is a major driver of phenotypic variation, and that considering the known geographic distribution of allelic variation across the range improves the match of model predictions to phenotypes expressed in natural populations. Specifically, variation in dormancy generated in the previous generation is predicted to cause life-cycle differences within a location, and the geographic distribution of allelic variation in dormancy interacts with local climatic environments to canalize an annual life history across the range. Finally, I test if allelic and environmental variation that affects early life stages can influence the environment experienced during reproduction. This environment determines both the time available for reproduction and the environment experienced during senescence. By implementing simple survival rules for flowering plants in the model, I show that time available for a plant to reproduce depends on earlier phenological traits and varies widely from year to year, location to location, and genotype to genotype. If reproductive trade-offs that underlie the evolution of senescence are environmentally sensitive, these results suggest that genetic variation in earlier life-stage transitions might shape senescence rates and whether they are environmentally responsive. In sum, my dissertation demonstrates the importance of pleiotropy, environment-dependent allelic expression, and epistasis in defining life-cycle variation, and proposes a novel way of predicting these relationships and complex life cycles under seasonal conditions.

Adaptation is central to our understanding of the origin of biological diversity. Yet whether adaptive divergence promotes the formation of new lineages remains poorly understood. My dissertation addresses the role of adaptive divergence in diversification and speciation. I also investigate an alternative mechanism: dispersal, which can promote diversification and speciation through its effects on gene flow and allopatry. To address the role of divergent adaptation and dispersal in the process of diversification, I take an integrated approach, combining both comparative methods with quantitative genetics to characterize patterns of diversification and speciation in the tribe Brassiceae and genus Cakile. I start with a comparative study of the role of dispersal and adaptation in diversification, and then focus on the role of climatic and latitudinal divergence in the processes of adaptive divergence and speciation. In general, I find limited evidence for the role of divergent adaptation in the evolution of intrinsic reproductive isolation. Diversification in the tribe Brassiceae appears to be mediated by dispersal ability, while in the genus Cakile, the evolution of intrinsic reproductive isolation is largely independent of ecological divergence. Thus, while divergent adaptation to novel habitats and climate are likely occurring in Brassiceae, mediated in part by the evolution of long-distance dispersal, the evolution of intrinsic genic reproductive barriers appears to not be influenced by adaptation.