Genome-wide association identifies adaptive refuge in Patagonia

M. FASANELLA; C.P. SOUTO; T. KITZBERGER; A.C. PREMOLI

Virtual

Congreso; III Virtual Meeting of Systematics, Biogeography, and Evolution.; 2022

SBE

Tree mortality caused by severe droughts has been reported worldwide. Drought is a major limitation for establishment, development, growth and survival in plants, especially in slow-growing species. Plants can react to changing environmental conditions through a combination of adaptation, phenotypic plasticity and migration, reducing extinction risks. In Patagonian transition forests one conifer Austrocedrus chilensis , hereafter Austrocedrus, dominates dry populations growing under stressful conditions, and is also found in mixed humid forests. We used Single Nucleotide Polymorphisms (SNPs) to test the hypothesis that dry-most populations hold adaptations to drought that are maintained by selection and are of great value for the persistence of their populations particularly under projected climate change scenarios of increased drought and fire frequency and extent.Ongoing studies from our research team showed that 92% of fire surviving trees are found in rocky outcrops at the dry-most end and that seedlings from Patagonian humid forest (HF) outgrow those from dry forest (DF) under common garden conditions suggesting genetically based differences. We sampled fresh leaves from 45 DF and 30 HF individuals along a west-east steep precipitation gradient within a latitudinal belt between 40 and 42°S to investigate the molecular genetics of drought tolerance in Austrocedrus . We extracted total genomic DNA and performed high-throughput GBS. After removing SNPs with low allele frequency (MAF>0.03), we kept 68 individuals with 3266 loci. We tested for outlier SNPs showing higher FST than neutral loci, using GenAlEx v6.5, OutFLANK and Pcadapt R packages. We implemented a Bayesian model–based clustering approach in STRUCTURE v.2.3 to analyze potential genetic differences between DF and HF, for both groups of SNPs (outliers and neutral). We implemented a PCA with 70 climate and soil variables to detect Genotype–Environment Associations (GEA) using: linear mixed model regressions (MLM) implemented in GAPIT, latent factor mixed models implemented in LFMM, and a Bayesian approach implemented in Bayenv2. SNP functional annotations were obtained blasting in TreeGenes, and NCBI databases.We obtained 24 outlier SNPs that differ between DF and HF (FST=0.118 p=0.001), whereas neutral genes (3242 SNPs) yielded no statistically significant divergence between DF and HF (FST=0.01 p=0.295). STRUCTURE results for adaptive SNPS showed two K groups that clearly separated HF individuals from DF ones, while a random assignment of ancestry to four different K clusters was obtained for neutral loci. GEA identified four marker-climate statistically significant associations driving genetic differentiation. Climate variables best explaining differentiation between DF and HF for outlier genes involved in leaf development and drought tolerance were: precipitation of wettest quarter, precipitation of coldest quarter, MAP, annual heat-moisture index, and summer precipitation.Observed genomic patterns for outlier loci and the lack of it for neutral ones suggest that despite gene flow depicted by the latter, natural selection would foster divergence for adaptive genes. Our study points to dry forests as refuges for drought-tolerant genetic variants adapted to cope with forecasted climate change and thus reducing extinction risks which is occurring faster than expected. So, these adaptive genes could provide potential drought adaptations to the species. Plants have developed different mechanisms to help them to handle abiotic stress, and we are beginning to grasping them. Our results contribute to understand the genomic basis of drought tolerance in conifers and provide information on adaptation that can be used to guide management measures and restoration efforts in the context of forecasted droughts.