Evolution of reproductive mode in the scaly-winged bark louse Echmepteryx hageni.

摘要

The origin and maintenance of sexual reproduction has long been of interest to evolutionary biologists. While the relationship between genetic diversity and asexuality has received recent attention, the precise nature of this relationship is dependent on the physiological mechanism (e.g. automixis vs. apomixes) and the details of the origin (e.g. endosymbiont infection or hybridogenesis) of asexual reproduction. Most of the work explicitly exploring the links between reproductive mode and genetic diversity has been theoretical in nature and has only modeled asexuality as clonal reproduction. Thus, the applicability of these studies beyond apomictic species is questionable. Empirical studies are necessary to understand the full breadth of the relationship, and ideal systems to do so are species with sexual and asexual forms co-existing. A common pattern of co-existence of reproductive mode is geographic parthenogenesis. This pattern is characterized by sexual populations occupying the central part of the species' range, while asexual populations are found in peripheral or marginal habitats. There is an alternative pattern of sexual-asexual co-existence that is present in several species of bark lice (Insecta: Psocoptera), an inconspicuous and under-studied group of insects that scrape lichen and other organic matter off a substrate, often bark, dead branches, rock outcroppings, or leaf litter. It is in many ways the inverse of geographic parthenogenesis. The asexual form of these species occupy the central part of the range, while sexual populations are restricted to peripheral or highly isolated localities. One such species is the scaly winged bark louse, Echmepteryx hageni, which is found through eastern North America, but sexual populations are known only from a few rock outcroppings in southern Illinois and eastern Kentucky. There is a strong disparity in mitochondrial genetic diversity between the reproductive forms of E. hageni, with the parthenogenetic form exhibiting very high haplotypic and nucleotide diversity while the sexual populations are virtually homogenous. In Chapter 1, I show that the disparity in genetic diversity is also present at nuclear loci, though not as strongly as at the mitochondrial markers. A demographic signal of rapid range expansion in the asexual form of E. hageni previously observed in mitochondrial data is also present at three of four nuclear loci. However, there are contrasting patterns of genetic structure and population differentiation between the two genomic regions. Finally, I outline three hypotheses that might explain the disparity in genetic diversity between the sexual and asexual forms present in the mitochondrial (and to a lesser extent the nuclear) genetic data: (1) sexuals are derived from the asexuals and are recent in origin, (2) asexual E. hageni have a greater per-year mutation rate due to life history differences, (3) conversion of a genetically diverse sexual species to a primarily asexual species. I test the first and third hypotheses in Chapter 2, both of which involve the question of the direction of evolution in reproductive mode in E. hageni. Hypothesis 1 is based on sexuality being a recently derived trait in E. hageni, while Hypothesis 3 posits a recent conversion of the species from sexual to primarily asexual reproduction. The derived position of sexual E. hageni observed in maximum parsimony analyses of mtDNA (Shreve et al 2011) is confirmed by maximum likelihood analysis, though with different root placement. However, outgroup randomization tests cast doubt on the ability of phylogenetic outgroups to accurately root the E. hageni tree. Once the evolutionary relatedness of mitochondrial haplotypes was taken into account via nesting of haplotype groups, measures of cytonuclear disequilibria at the four nuclear loci are consistent with the relatively recent conversion of a diverse, sexual species to parthenogenetic reproduction, consistent with Hypothesis 3. However, a secondary assumption of Hypothesis 3, that parthenogenesis is the result of an endosymbiont infection, is not supported. Both Wolbachia and Rickettsia are found in some E. hageni, but infection does not appear to be correlated with asexual reproduction. Chapter 3 compares patterns of mitochondrial genetic diversity in E. hageni with another bark louse species, Peripsocus subfasciatus, in order to test Hypothesis 2. The multivoltine sexual form of P. subfasciatus has significantly greater haplotypic and nucleotide diversity than the univoline sexual E. hageni. In addition, unlike E. hageni, the genetic diversity of sexual P. subfasciatus is slightly but statistically greater than in the asexual form of the species. These results are consistent with the hypothesis that life history differences between the multivoltine asexual form and the univoltine sexual form of E. hageni are at least partially responsible for the disparity in genetic diversity between the two reproductive forms of this species. Asexual genetic diversity is elevated in P. subfasciatus in a manner similar to E. hageni, suggesting that this may be a common pattern in the restricted-sexuality distribution. However, additional studies of other bark lice with this pattern are necessary to verify the trend. Chapter 4 examines the extreme range disjunctions present in P. subfasciatus from a phylogeographic and population genetic perspective. P. subfasciatus is found in eastern North America, the Pacific coast of the western North America, and also in Europe. In addition, sexuality is restricted to western North America and isolated populations in eastern North America. There are two mitochondrial clades with a mean sequence divergence between them of 1.8%, but this genetic break does not correspond to reproductive mode or to geography. In fact, reproductive mode has virtually no association with genetic structure at all. On the contrary, analysis of mitochondrial and nuclear data finds that grouping western North American and European populations versus eastern North American populations explains more of the genetic variation within P. subfasciatus than any other grouping scheme. European and western North American population may therefore have historically connected via the Bering land bridge, but phylogeographic history of the eastern North American-European split remains unclear. The experiments and analyses carried out as part of the dissertation have variously both supported and contradicted each of the three hypotheses. As a result, the evolutionary processes shaping genetic diversity and distributional patterns in E. hageni remain unclear. The three hypotheses are not all mutually exclusive, and it is possible that life history differences have worked in conjunction with other evolutionary and ecological processes to shape genetic diversity within E. hageni. One plausible sequence of events is that a rapid conversion from sexual to asexual reproduction in E. hageni resulted in a majority of the standing genetic diversity being preserved in the parthenogenetic lineages. At this point, the life history differences between the sexual and asexual forms and their differences in effective population size could have worked together to maintain and exacerbate the differences in genetic diversity between the reproductive forms.