Seminars

November 17 2016

5:00 pm TLSB 1100

EcoEvoPub Series

Graduate Student Presentations

Summary

Devaughn Fraser
Department of Ecology and Evolutionary Biology, UCLA

“Effect of anticoagulant rodenticides on transcriptional profiles in bobcats (Lynx rufus)”

Anticoagulant rodenticides (ARs) are the primary chemical control method employed to suppress rodent populations worldwide. AR exposure, however, poses a major conservation threat to non-target species, especially carnivores and raptors for which rodents are a staple food item. While direct mortality from secondary exposure is the most well documented concern for wild populations, accumulating evidence suggest that chronic, sub-lethal exposure can also have negative impacts at the individual and population level. For example, increased mange-susceptibility in bobcats utilizing peri-urban habitats of the Los Angeles area likely results from severe immune dysregulation caused by AR exposure, leading to high disease-related mortality and driving a genetic bottleneck in a local population. To elucidate potential molecular mechanisms underlying this observation, we conducted RNA-seq analyses on whole blood samples from free-ranging bobcats captured in the Santa Monica Mountains, for which AR exposure was (n = 26) and was not (n = 26) detected in whole blood. We used a linear mixed model to assess the effect of exposure on genome-wide expression in whole blood while controlling for technical effects introduced during sample handling. Our models identified several genes directly involved in the endoplasmic reticulum stress response, toxin metabolism, blood coagulation and immune function.  GO enrichment analysis showed down-regulated genes were highly represented by terms related to platelet ctivation, wound healing and inflammation; while up-regulated genes were enriched for terms associated predominantly with T-cell mediated immune response, and specifically a Th1 immune profile. Additionally, we performed a transcript origin analysis (TOA) to infer the specific cellular mediator of AR effects. Our results show that AR exposure is associated with an overall decrease in transcripts associated with monocytes (particularly immature Ly6c-high/CD16- monocytes) and naive B cells, and an increase in transcripts associated with memory B-cell lymphocytes. Collectively, our results suggest that chronic AR exposure alters both innate and adaptive immune capabilities, limiting the ability of the animal to mount an appropriate immune response against certain infections. Therefore, ARs may pose a substantial threat to small populations facing additional threats such as disease. To our knowledge this RNA-seq study is one of the first that investigate the effect of anticoagulant rodenticides on a wild species and may set an important precedent in future research by demonstrating the utility of RNA-seq to study the impacts of chemical stressors on natural populations.

Charlie De La Rosa

Department of Ecology and Evolutionary Biology, UCLA

“Cows that browse: foraging ecology of a domestic ungulate at different scales”

Cattle ranching is one of humankind’s oldest and most widespread means of extracting sustenance and economic value from large areas of undeveloped land.  Domestic cattle (Bos taurus) are grazing ruminants, preferring to forage on graminoids, forbs, and other relatively high-nutrient value vegetation.  In predominantly forested ecosystems, however, they become facultative browsers, feeding on a wide range of woody plant and succulent stems, leaves, and fruits, in addition to preferred vegetation types.  Although there are compelling conservation payoffs to predicting cattle foraging preferences on woody vegetation in forests, the low nutritional value, potential for toxicity, and spatial heterogeneity of these plants complicate existing foraging theory models.  As a result, few researchers have tested foraging and behavioral theories of this type in empirical settings.
I developed an animal-mounted data collection system and plant census protocols to study free-ranging cattle foraging and ranging in tropical deciduous forest (TDF) in the Alamos Municipality of southeastern Sonora, Mexico.  I then used an information-theoretic statistical approach to address questions about dietary selectivity in cows at different scales.  At the ranch-level (100+ hectare) scale, the best predictor of frequency in diet for most woody plants is frequency in the environment, although some plants are avoided.  The pattern is reinforced at the foraging station (25m2) scale.  My preliminary results suggest that cows are not picky eaters—overall, they eat what is abundant, but they will select some species and avoid others, given the choices in front of them.  

 

 

 

 

 



 

 

 

 

 

 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



this is idtest: