Evolutionary quantitative genomics ofPopulus trichocarpa

Abstract

Forest trees generally show high levels of local adaptation and efforts focusing on understanding adaptation to climate will be crucial for species survival and management. Here, we address fundamental questions regarding the molecular basis of adaptation in undomesticated forest tree populations to past climatic environments by employing an integrative quantitative genetics and landscape genomics approach. Using this comprehensive approach, we studied the molecular basis of climate adaptation in 433 Populus trichocarpa (black cottonwood) genotypes originating across western North America. Variation in 74 field-assessed traits (growth, ecophysiology, phenology, leaf stomata, wood, and disease resistance) was investigated for signatures of selection (comparing Q ST -F ST ) using clustering of individuals by climate of origin (temperature and precipitation). 29,354 SNPs were investigated employing three different outlier detection methods and marker-inferred relatedness was estimated to obtain the narrow-sense estimate of population differentiation in wild populations. In addition, we compared our results with previously assessed selection of candidate SNPs using the 25 topographical units (drainages) across the P. trichocarpa sampling range as population groupings. Narrow-sense Q ST for 53% of distinct field traits was significantly divergent from expectations of neutrality (indicating adaptive trait variation); 2,855 SNPs showed signals of diversifying selection and of these, 118 SNPs (within 81 genes) were associated with adaptive traits (based on significant Q ST ). Many SNPs were putatively pleiotropic for functionally uncorrelated adaptive traits, such as autumn phenology, height, and disease resistance. Evolutionary quantitative genomics in P. trichocarpa provides an enhanced understanding regarding the molecular basis of climate-driven selection in forest trees and we highlight that important loci underlying adaptive trait variation also show relationship to climate of origin. We consider our approach the most comprehensive, PLOS ONE |

Genomic and functional approaches reveal a case of adaptive introgression fromPopulus balsamifera(balsam poplar) inP. trichocarpa(black cottonwood)

“…Admixed individuals with these introgressed P. balsamifera haplotypes are restricted to northern and central areas of the geographical range of P. trichocarpa, and have been assigned to distinct climate clusters (Porth et al . ). Based on historical climate data, northern and north‐central regions have significantly lower mean annual temperature (MAT: 4.2 °C), number of frost‐free days (NFFD: 175 days), and mean annual precipitation (MAP: 744 mm) than coastal and southern regions (MAT: 9.5 °C, NFFD: 287 days, MAP 2805 mm) (Porth et al .…”

“…Based on historical climate data, northern and north‐central regions have significantly lower mean annual temperature (MAT: 4.2 °C), number of frost‐free days (NFFD: 175 days), and mean annual precipitation (MAP: 744 mm) than coastal and southern regions (MAT: 9.5 °C, NFFD: 287 days, MAP 2805 mm) (Porth et al . ), suggesting potential niche differentiation between populations with and without the introgressed region.…”

Introgression fromPopulus balsamiferaunderlies adaptively significant variation and range boundaries inP. trichocarpa

“…Two orthologs of the nitrate transporter AtNRT2 (Potri.009G008500, Potri.009G008600), from the Major Facilitator Superfamily (PF07690), were associated with a number of phenological (bud set, yellowing, leaf drop, growing period, height growth cessation) and biomass traits (height, height gain) here and in a previous association analysis (McKown et al ., ). One of the AtNRT2 orthologs (Potri.009G008600), which was also associated with damage by disease in a different association study (La Mantia et al ., ), showed exceptionally strong correlations with geoclimate variables including latitude and temperature (Geraldes et al ., ; Porth et al ., ), and had SNPs that were F ST outliers in a landscape genomic analysis (Geraldes et al ., ). These previous reports used the same P. trichocarpa populations but a different genotyping method and association mapping approach than those used in the present study.…”

Independent evolution of ab- and adaxial stomatal density enables adaptation

“…Within Populus, populations are more am-227 phistomatous at Northern latitudes with shorter growing seasons that may select for 228 faster carbon assimilation (McKown et al, 2014; Kaluthota et al, 2015). Latitudinal 229 variation Populus trichocarpa is due mostly to adaptive variation in adaxial stomatal 230 density (McKown et al, 2014;Porth et al, 2015). Stomatal density rather than size may have responded more readily to selection because there is no genetic covariance 232 between ab-and adaxial stomatal density, permitting independent evolution(Porth 233 et al, 2015).…”