Utilization of Genetic and Genomic Resources in Hardwood Forest Trees

This thesis addresses three research projects involving forest tree genetics and genomics. The first (Chapter 2) examined mechanisms for the mixed genetic ancestry in many of the 48 populations included in a range-wide genetic study of green ash (Fraxinus pennsylvanica Marshall), a species under acute threat from the invasive phloem feeding insect, the emerald ash borer (Agrilius planipennis Fairmaire). Emerald ash borer (EAB) threatens all 16 of the ash species in North America but the threat is most acute for green ash, the most preferred host and one of the most widely distributed trees in this region (Cappaert et al. 2005; MacFarlane and Patterson Meyer 2005). I hypothesized that because green ash was extensively used for shelter belts in rural areas, restoration of riparian zones, and landscaping trees in urban areas, the admixture detected could be signs of human-assisted migration and introgression from planted cultivars. To test this hypothesis, I genotyped green ash cultivars to look for evidence of cultivar introgression into natural populations. My work demonstrated that gene flow can occur from mass-distributed tree cultivars into naturally regenerated forests.

For the second project (Chapter 3), I conducted a bait-capture target enrichment using species of ash (Fraxinus, Oleaceae), oak (Quercus, Fagaceae), and chestnut (Castanea, Fagaceae). For this experiment we targeted EST-SSR microsatellites-containing sequences from the Fagaceae, the family containing oak and chestnut, and the Oleaceae, the family containing ash. This project demonstrated how one can reduce time-consuming marker development with a massive screening of markers and demonstrated the feasibility of using this technology in forest trees effectively and efficiently.

For the final project (Chapter 4), I used previously sequenced ddRADseq data to generate a high-density linkage map in black walnut (Juglans nigra L.). This map was constructed with a full-sib mapping population (i.e. a group of progeny from the same two parents) and includes 473 ddRADseq markers (SNP polymorphisms) with an average distance between markers of ~4.33 cM. All Juglans species have 16 haploid chromosomes (n=16) and our map has the expected 16 linkage groups. This map will be further refined using EST-SSR markers from the transcriptomes of black walnut and Persian walnut (J. regia L.). These markers will be tested for mapping informative polymorphisms and the mapping population genotyped using the same target-capture technologies utilized for the second project (Chapter 3). This linkage map will be used to examine synteny and collinearity in Juglans species, for QTL mapping in black walnut, and for enhancing the genome assembly of Juglans species sequenced to date (Stevens et al. 2018).

My projects reveal how, even when working with challenging non-model species with long juvenile periods and complicated genomes, it is possible to answer important basic genetics and genomics questions while generating data that is immediately useful for conservation and protection of threatened forest tree species. We show how wise experimental design and creative use of recent technologies can accomplish this dual purpose even with a comparatively small amount of funding.