A cellular basis for endosymbiont prevalence


Endosymbioses are the most intimate form of species interaction, as microbes reside within the cells of their host. My research focuses on the spread and evolution of Wolbachia bacteria, the most common endosymbionts in nature. Wolbachia infect about half of all insects species and biocontrol programs use virus-blocking Wolbachia from Drosophila melanogaster to block transmission of deadly human diseases like dengue and Zika in mosquitoes. I take an integrative approach to examine how genomic and cellular host-microbe interactions dictate global patterns of Wolbachia prevalence and (co)evolution with insect hosts.

Transmission rates underlie Wolbachia prevalence

The evolution of endosymbioses fundamentally depends on efficient microbe transmission to host offspring. Our work and others' has shown that Wolbachia prevalence (the frequency of infected hosts) can vary widely among host populations, including in Australia. We recently made the exciting discovery that cold temperatures perturb the cellular basis for Wolbachia vertical-maternal transmission during host oogenesis, explaining continent-wide clines of Wolbachia prevalence. Decoupling temperate-evolved Wolbachia and hosts further perturbs transmission in the cold, implying rapid co-adaptation to climate in only the last ~5,000 years. My genomic analyses indicate that Wolbachia maternal transmission is mediated in part by host interactions with the major Wolbachia surface protein WspB. This new research has important implications for Wolbachia-based biocontrol programs, because very little is known about the function of any Wolbachia loci and biocontrol requires efficient Wolbachia transmission in the face of stressful thermal conditions.


Endoymbiont effects on host fitness

Microbes have diverse effects on host physiology, behavior, and fitness. Wolbachia reside throughout host somatic tissue, including the brain, yet we have a poor understanding of how Wolbachia influence components of host fitness to rapidly spread in nature. By taking a comparative approach, we've found that divergent Wolbachia strains have pervasive effects on host physiology and behavior. Thus far, our work has demonstrated how Wolbachia alter host activity levels and thermal preference. Chelsey Caldwell, a UM undergraduate student, spearheaded the project on host thermal preference and presented her findings virtually at the UM Conference on Undergraduate Research!

Evolution at the interface of toxin-binding in a predator-prey arms race

Screen Shot 2019-06-10 at 3.25.44 PM.png

Intense reciprocal selection between natural enemies like predator and prey can lead to arms race coevolution. ​I study how constraints shape evolution at the molecular interface of toxin-binding in the coevolutionary arms race between resistant garter snakes and their toxic prey, Pacific newts. My dissertation in the Brodie lab examined how protein evolution, gene flow, and historical biogeography gave rise to geographic variation in arms race dynamics between predator and prey

Convergent evolution of a conserved protein

Two separate lineages of the common garter snake have convergently evolved resistance to tetrodotoxin (TTX), the deadly neurotoxin found in newts. We found that both snake lineages independently evolved resistance through a common series of amino acid changes to the Nav1.4 skeletal muscle sodium channel that disrupt toxin binding. In each instance, evolution proceeded through the same first-step mutation to the channel pore, suggesting the initial change had permissive effects for subsequent increases in resistance.

We further found that once TTX-resistant mutations accumulated in Nav1.4, negative trade-offs arose with sodium channel function and muscle performance. Our results indicate that costs develop as a consequence of accumulating mutations in the arms race that beneficially interfere with toxin binding. The evolutionary trajectory of Nav1.4 strikes a balance between TTX resistance and the maintenance of conserved channel function. More broadly, balancing selection and pleiotropy associated with Nav1.4 seem to explain geographic variation in snake TTX resistance across western North America.

Screen Shot 2019-06-11 at 1.16.09 PM.png

Asymmetries in the arms race

Reciprocal selection is the hallmark of a coevolutionary relationship, but the evolutionary response of each species may not be symmetrical. We tested whether snakes and newts exhibit symmetrical evidence of local co-adaptation across their range of sympatry. We found that levels of snake resistance and newt toxins are closely matched across the landscape, implying that mosaic variation in the armaments of both species is a result of local pressures in the arms race. By the same token, phenotypic and genetic variation in snake resistance exhibit a clear signature of local adaptation. Contrary to conventional wisdom, however, landscape variation in newt toxins was best predicted by population genetic structure rather than predator resistance. Newt populations seem to structure variation in prey toxin levels, which in turn influences local levels of resistance in predators. This unexpected result implies that neutral processes like gene flowrather than reciprocal selectionmay represent the greatest source of variation across the geographic mosaic of coevolution.


Patterns of genetic diversity in sympatric lizards

Screen Shot 2019-06-11 at 1.25.48 PM.png

Genetic variation is a fundamental requirement for evolution and adaptation. Genetic diversity should increase with effective population size and the decreasing effects of drift. For my Master's thesis in the Routman Lab, I tested how levels of genetic diversity vary in sympatric populations of lizard species in the Mojave Dessert that differ in population size and other ecological factors. At both mitochondrial and nuclear loci, we found that abundant, short-lived species had significantly higher estimates of diversity than less abundant, long-lived lizards. Higher diversity in the abundant species is likely due to a number of demographic factors, including larger effective population sizes, short generation times, and high rates of gene flow from surrounding populations.

Screen Shot 2019-06-11 at 2.47.40 PM.png