Phylogeography, reproductive isolation, and the evolution of sex determination mechanisms in the copepod Tigriopus californicus

2018-09-22The processes and patterns underlying speciation are a major focus of evolutionary biology. These processes often result in the evolution of reproductive isolation, and sex chromosomes often play an important role in postzygotic reproductive isolation. However, sex chromosomes—and, more generally, sex determination mechanisms—are themselves highly evolutionarily labile, and many lineages lack sex chromosomes but still manage to speciate. However, the vast majority of work on the evolution of postzygotic reproductive isolation has been conducted in taxa that do possess sex chromosomes, leaving many open questions regarding how postzygotic isolation may evolve in taxa without sex chromosomes. This dissertation examines how the speciation process may differ in taxa that do not have sex chromosomes. ❧ Chapter 1 asks whether the presence of sex chromosomes in a clade accelerates speciation. Sex chromosomes disproportionately accumulate postzygotic incompatibilities (large X–effect) and contribute to sex–specific hybrid inviability and infertility (Haldane’s rule). Therefore, in the absence of sex chromosomes, postzygotic reproductive isolation may take longer to accumulate to the same level, which could ultimately result in slower speciation rates. We addressed this question in reptiles, where we found that speciation intervals are shorter in snakes and lizards (where sex chromosomes are common) than in turtles and crocodiles (where sex chromosomes are rare or absent). ❧ Chapters 2–4 examine phylogeography, hybrid incompatibility, and sex determination in the intertidal copepod Tigriopus californicus, a species with polygenic (genetic but non-chromosomal) sex determination. Chapter 2 presents a phylogeographic study of T. californicus across the species range. T. californicus populations have been documented from southern Alaska, USA to central Baja California, Mexico, and previous phylogeographic studies have employed only a single mitochondrial locus (mtCOI). We used a different mitochondrial locus (mt12S), as well as 7 nuclear loci (18S, ITS1, 5.8S, and 4 transcriptome–based loci) to resolve relationships between major clades and to clarify the taxonomic status of populations at the southern edge of the range, which are genetically and reproductively isolated from other T. californicus populations. We found definitive evidence that these southern populations constitute a novel species that is geographically continuous with T. californicus, identify the presence of major regional clades, and document a general isolation–by–distance pattern within regions. ❧ In Chapter 3, non-recombinant backcrosses between multiple pairs of T. californicus populations were used to map postzygotic reproductive isolation, morphometric features, and sex. We identified segregation distortion in some, but not all crosses and detected few epistatic incompatibilities. We mapped associations with sex to several chromosomes, consistent with polygenic sex determination, and we found evidence of transgressive segregation for morphometric traits in some crosses. In these crosses, we found that males consistently fare worse than females, despite not being heterogametic (the traditional explanations for sex-specific deleterious effects invoke heterogamety). Thus, explanations for sex-specific deleterious effects of hybridization may require the generation of alternative hypotheses in taxa without sex chromosomes. ❧ Finally, Chapter 4 investigates polygenic sex determination and the emergence of a novel sex chromosome in a hybrid population of T. californicus. Polygenic sex determination, where sex determination is genetic but non-chromosomal, is predicted to be unstable and prone to takeover by strong sex-determiners, which can then become sex chromosomes. We found that alleles from one parental population (SD, from San Diego, California, USA) had introgressed sex–specifically on one chromosome (chromosome 10), despite swamping by the other population’s alleles (SC, from Santa Cruz, California, USA) across the rest of the genome. SD alleles on chromosome 10 appear to have a strong feminizing effect, and—in the absence of meiotic recombination in females—to have been transmitted almost exclusively to females. Whole-genome re-sequencing of the parental progenitor populations did not reveal allele frequency differences between males and females to be concentrated anywhere in the genome, whereas in the hybrid population these differences were elevated on chromosome 10. The total evidence suggests that chromosome 10 is acting as a novel sex chromosome within the hybrid population but not within the parental populations, making it perhaps the first sex chromosome to emerge from a polygenic sex determination system in a laboratory. ❧ The speciation process is currently poorly understood outside of a few model taxa, where sex chromosomes are involved in most aspects of postzygotic reproductive isolation. Increasing the taxonomic breadth of this research is an imperative in the Anthropocene, where human activities are directly affecting the speciation process through anthropogenically–mediated hybridization and extinction. Speciation in the absence of sex chromosomes may proceed fundamentally differently, as depicted in Chapters 1, 2, and 3. Alternatively, as seen in Chapter 4, hybridization could in some cases lead to the emergence of a novel sex chromosome, which could then modify the speciation process