Date of Degree


Document Type


Degree Name





Anthony Bruce Wilson

Committee Members

Gregory O’Mullan

Susan Elizabeth Alter

Robert Anderson

Nolwenn Dheilly

Subject Categories

Ecology and Evolutionary Biology | Molecular Genetics


Adaptive immunity, major histocompatibility, microbiome, local adaptation, population genetics, New Zealand


Local adaptation is a fundamental concept at the core of evolutionary process. Devising means to break down larger concepts and systematically test internal mechanisms are key to understanding the driving forces behind the influence of environmental differences on biological systems. To understand dynamic processes in the natural world, we must strip away layers and confounding variables that may mask the signal we hope to understand. In this case seahorse species, Hippocampus abdominalis, or the pot-bellied seahorse, made an unlikely but ideal study system. The first layer to strip away was neutral genetic variation within the population. A multilocus genetic dataset composed of the mitochondrial control region and nuclear microsatellites revealed population structure throughout the species’ geographical range. Australian H. abdominalis colonized the South Island of New Zealand during the previous interglacial-glacial cycle between twelve and 120 Ka followed by population expansion and migration northward to the North Island of New Zealand. While the region is highly diverse with strong evidence of ancestral and contemporary connectivity, Australian populations demonstrated higher geographic structure, lower diversity and lower connectivity than New Zealand populations. Contemporary migration via rafting is thought to be responsible for low but significant dispersal in the direction of New Zealand to Australia. All contemporary population sizes were smaller than ancestral estimates throughout the range, suggesting the potential loss of rare alleles over time. By comparison, immune gene locus major histocompatibility class II alpha and beta (MHIIab) had a similar level of diversity to neutral nuclear microsatellites with contrasting patterns of distribution across the species range. Specific MHIIab alleles correlated highly with particular microbes identified in the gut tract, which were derived from the environment. While gut microbial communities significantly differed by environmental types defined by terrestrial wastewater contributions from human activity (or a lack thereof), overall MHIIab population structure did not demonstrate the same pattern. Heterozygote advantage appeared to yield a stronger signal due to observed heterozygote excess at this locus than selection pressure from any particular microbe or microbial community. A subset of these microbial-host correlations within environmental types, however, followed expected patterns. Candidate microbes were selected for controlled exposure trials that were associated with alleles specific to urban human-associated nearshore environments (Photobacterium phosphoreum) or rural pristine-associated nearshore environments remote from any human activity including agriculture or aquaculture (Staphylococcus epidermidis). Supertyping analyses of all potentially expressed MHIIab cell surface proteins from wild populations identified 17 functionally similar gene clusters. Experimental populations of H. abdominalis fry were bred from captive adults carrying supertypes containing MHIIab alleles correlated with either of the two microbes. Overall, fitness and survival were higher after pristine-associated microbial exposure and lower after human-associated microbial exposure for the twenty genotypes vs. supertypes tested relative to the total range of growth observed over time. Two genotypes (one from each associated exposure condition) demonstrated increased fitness with one microbe over the other, and three genotypes (from the human-associated exposure condition) demonstrated decreased fitness. Results from the mixed-effects treatment were highly variable among the genotypes, supporting the paradigm that these microbial-host relationships are influenced by microbe community and concentration in the natural environment. To further isolate important relationships and selection pressures between differing environments, these investigations can be repeated, additional microbes and genotypes can be tested, and eventually in situ transplant experiments can be conducted. Hopefully this research inspires further studies of non-model organisms or novel systems that can look at important scientific questions from new or different angles, yielding transformative insights into mechanisms driving local adaptation.