The majority of flowering plant species rely on animals to pollinate their flowers, and in most cases such pollination results in cross-fertilization (i.e., mating between different individuals). However, nearly 25% of species regularly self-fertilize, with pollen being transferred within or between flowers on the same plant. Within natural populations, self-fertilization causes a reduction in genetic variation within individuals and their descendents, in a similar way to the effects of inbreeding in cultivated crops. Given that there are many potential ecological and evolutionary disadvantages to this loss of genetic variation, the evolution of self-fertilization in many plant species remains enigmatic. To date, most explanations for the phenomenon have proposed reproductive advantages that could outweigh the known genetic disadvantages. For example, natural selection might favor self-fertilization when pollinators are scarce or unreliable, or when short growing seasons favor rapid reproduction. However, selfing often evolves along with a suite of physiological, morphological, and life history traits, which raises the alternative possibility that its evolution is influenced by selection on other traits with which it is developmentally, physiologically, or genetically correlated. The goal of this research is to seek evidence for that alternative explanation. Specifically, using a combination of quantitative genetic and physiological approaches, the investigators will test the hypothesis that selfing can evolve as a consequence of genetic correlations between (a) floral traits that affect the rate of self-fertilization and (b) life history or physiological traits that enable plants to escape drought. If selection on life history or physiological traits does influence selfing rates, this will have cascading effects on the genetic structure of populations, potentially limiting their ability to adapt to future environmental change. The annual wildflower genus Clarkia (Onagraceae) provides a rich opportunity to investigate such effects because it includes several species in which self-fertilization has evolved independently.
Broader impacts: This project will initiate a collaboration and reciprocal training among four PIs and include the participation of at least 15 undergraduates, including those of under-represented groups. PIs will work with the University of California, Santa Barbara's "Kids in Nature" program to integrate research and education. In addition, physiological comparisons between species and subspecies may generate predictions about changes in species' distributions and the genetic risks associated with inbreeding in the face of an increasing frequency of droughts that the southern Sierra Nevada is expected to experience in coming years.
Most species of flowering plants (angiosperms) rely on animals to pollinate them, which usually results in mating between different individuals (outcrossing). Nearly 25% of angiosperm species, however, regularly self-fertilize, whereby a plant's pollen fertilizes its own flowers. Relative to outcrossing, self-fertilization reduces genetic variation among a plant’s descendants and often results in poorer plant performance due to the effects of inbreeding. The evolution of self-fertilization from outcrossing ancestors may also reduce a population’s ability to adapt to environmental change due to the gradual loss of genetic diversity. Most explanations for the evolution of self-fertilization propose that its advantages — such as the ability to set seed where pollinators are scarce or unreliable — outweigh the known disadvantages. However, selfing often evolves along with a suite of other traits, which raises the possibility that selfing may evolve in part (or even largely) due to selection on other traits that are strongly genetically correlated with self-fertilization. In this case, if natural selection in wild populations favors individuals with life history or physiological traits that are genetically associated with floral traits that promote self-fertilization, then the tendency to self-fertilize may evolve as a correlated response to selection even if it has negative effects such as a reduction in plant performance or reduced genetic variation. The goal of this research was to test this alternative explanation for the evolution of self-fertilization by examining wild and experimental populations of four closely related species and subspecies in the annual wildflower genus Clarkia (Onagraceae). These species include the outcrossing species, C. unguiculata and its highly selfing closest relative, C. exilis, and two subspecies of Clarkia xantiana (the outcrossing subspecies xantiana and the self-fertilizing subspecies parviflora). By examining two pairs of taxa with contrasting mating systems, we could assess whether the consequences of the evolution of selfing are similar in both cases. Among these taxa, the selfers flower earlier and complete their life cycles earlier under both greenhouse and field conditions. In the current research, we observed physiological rates and plant performance in wild populations in the southern Sierra Nevada, and we carried out controlled greenhouse experiments, to test the hypothesis that genetic associations between floral traits that promote self-fertilization and life history or physiological traits that enable plants to escape drought may have contributed to the evolution of selfing in Clarkia. If this hypothesis were to be corroborated, then populations of annual plants consistently exposed to water stress in spring or summer would be likely to evolve faster life cycles and higher rates of self-fertilization, along with its associated risks. To date, we have found that photosynthetic and transpiration rates of selfing populations are faster than those of their outcrossing counterparts. This result is consistent with the hypothesis that natural selection has favored faster gas exchange rates in selfers than in their outcrossing sister taxa, potentially because faster physiological rates enable the former to achieve their earlier maturation and faster life cycles. Further support for this hypothesis comes from our observation that higher photosynthetic rates are more strongly associated with reproductive fitness (estimated as lifetime fruit production) in C. exilis than in C. unguiculata. Analyses are underway to determine whether genotypes and taxa that escape drought consistently exhibit lower water use efficiency, lower thermal energy dissipation, and lower anti-oxidant enzyme activity than their later-flowering counterparts. Detailed, multi-year studies of pollination in multiple populations of each of the outcrossing taxa found no evidence that selfing was advantageous due to a lower availability of pollinators early in the flowering season. Finally, when C. unguiculata plants that had been selectively bred for early-flowering were raised in the field, we detected no significant genetically-based association between flowering time and one crucial floral trait (close proximity of anthers and stigma) that promotes selfing. These results suggest that the combination of early flowering and self-fertilization in Clarkia exilis is probably due to their independent, rather than correlated, evolution. Broader impacts: This project initiated a collaboration among four PIs, two of whom are at a predominantly undergraduate institution. Field, greenhouse, and lab work included the participation of >40 undergraduates, including Asian American, Latin American, Native American, and African American students, many of whom are pursuing graduate programs in organismal biology or in health-related programs. One PhD student (now an Assistant Professor) relied on this project to facilitate her research. Participating undergraduates gained several of the following skills: interpreting the scientific literature; using a gas exchange analyzer; cultivating, pollinating and harvesting plants under greenhouse conditions; recording morphological, architectural, developmental, floral, life history and physiological traits; measuring anti-oxidant production; and data management. Data generated here are also being used to develop predictions about changes in species’ distributions and the potential genetic risks associated with inbreeding in the context of an increasing frequency or intensity of droughts that the southern Sierra Nevada may experience in the future.
