Molecular signatures of selection in Populus fremontii: Investigation of candidate outlier loci associated with environmental variation

Erica L. Sukovich, US Geological Survey, Flagstaff, AZ; Biology, Northern Arizona University, Flagstaff, AZ, Helen M. Bothwell, Research School of Biology, Australian National University, Canberra, ACT, Australia, Hillary F. Cooper, Center for Adaptable Western Landscapes, Northern Arizona University, Flagstaff, AZ, Robert T. Massatti, Southwest Biological Science Center, U.S. Geological Survey, Flagstaff, AZ, Amanda R. De La Torre, Forestry, Northern Arizona University, Flagstaff, AZ, Catherine Gehring, Merriam Powell Center for Environmental Research, Northern Arizona University, Flagstaff, AZ and Gerard J. Allan, Merriam-Powell Center for Environmental Research, Northern Arizona University, Flagstaff, AZ

Background/Question/Methods
As species experience increased warming and drying due to climate change, it is becoming increasingly important to understand the genetic basis of their adaptive capacity to environmental stress. This is especially true for forest trees whose distributions encompass a wide range of environmental variation to which populations typically become locally adapted. Local adaptation is significant because it is often closely linked to productivity, biodiversity and associated ecosystem services. In this study we investigated the molecular basis of adaptation to two key variables, precipitation and temperature, in Populus fremontii, a foundation forest tree. We asked: 1. To what extent is adaptation to temperature and precipitation detectable at the molecular level? 2. If detectable, are there loci that are more strongly associated with variation in precipitation or temperature? 3. If specific loci show strong associations with temperature and precipitation, what traits or physiological processes do they control? Using a RADseq dataset consisting of ~15,000 SNPs from 58 source populations and 447 individuals, we searched for outlier loci associated with 7 bio variables obtained from WorldClim and analyzed our results using Redundant Discriminant Analysis (RDA). Additionally, we used a gene ontology database to identify possible functions of outliers associated with specific variables.
Results/Conclusions
We found strong evidence for molecular-based adaptation in the form of 1,288 outlier loci, associated with variation in precipitation and temperature. Identified outliers were 3 times more likely to be associated with precipitation variables (BIO14, BIO18, and BIO19) than temperature variables (BIO2, BIO4, BIO9, and BIO11). The gene ontology analysis yielded a few functional proteins of interest, including an ethylene-responsive transcription factor that plays a role in cold, drought, and high salinity stress responses and NAC domain-containing protein 55, which is associated with drought tolerance. We also identified a significant relationship between three previously identified ecoregions and the environmental predictors. Sites located in the Utah High Plateau ecoregion had more outliers that were being driven by precipitation of the driest month (BIO14), while the Central California Valley ecoregion was driven by precipitation of the coldest quarter (BIO19). The Sonoran Desert ecoregion showed that outliers were being driven nearly equally by precipitation of the warmest quarter (BIO18) and mean temperature of the coldest quarter (BIO11). These findings are consistent with key environmental variables that likely drive local adaptation within the three ecoregions and provide insight into the molecular underpinnings of how Populus fremontii may adapt to a rapidly changing climate.