Abstract
Because earth is currently experiencing unprecedented climate change, it is important to predict how species will respond to it. However, geographically-explicit predictive studies frequently ignore that species are comprised of genetically diverse individuals that can vary in their degree of adaptation to extreme local environments; properties that will determine the species’ ability to withstand climate change. Because an increase in extreme drought events is expected to challenge plant communities with global warming, we carried out a greenhouse experiment to investigate which genetic variants predict surviving an extreme drought event and how those variants are distributed across Eurasian Arabidopsis thaliana individuals. Genetic variants conferring higher drought survival showed signatures of polygenic adaptation, and were more frequently found in Mediterranean and Scandinavian regions. Using geoenvironmental models, we predicted that Central European populations might lag behind in adaptation by the end of the 21st century. Further analyses showed that a population decline could nevertheless be compensated by natural selection acting efficiently over standing variation or by migration of adapted individuals from populations at the margins of the species’ distribution. These findings highlight the importance of within-species genetic heterogeneity in facilitating an evolutionary response to a changing climate.
One-sentence summary “Future genetic changes in A. thaliana populations can be forecast by combining climate change models with genomic predictions based on experimental phenotypic data.”
Ongoing climate change has already shifted latitudinal and altitudinal distributions of many plant species (1). Future changes in distributions by local extinctions and migrations are most commonly inferred from niche models that are based on current climate across species ranges (2, 3). Such approaches, however, ignore that an adaptive response can occur also in situ if there is sufficient variation in genes responsible for local adaptation (4–6). The plant Arabidopsis thaliana is found under a wide range of contrasting climates, making it distinctively suited to study evolutionary adaptation to a changing climate (7–9). For the next 50 to 100 years, it is predicted that extreme drought events, potentially one of the strongest climate change-related selective pressures (10), will become pervasive across the Eurasian range of A. thaliana (2, 11). An attractive hypothesis is that populations from the southern edge of the species’ range (12) provide a reservoir of genetic variants that can make individuals resistant to future, more extreme, climate conditions (12, 13). To investigate the potential of A. thaliana to adapt to extreme drought events, we first linked genetic variation to survival under an experimental extreme drought treatment (14–16). By combining genome-wide association (GWA) techniques that capture signals of local and/or polygenic adaptation (17, 18) with environmental niche models (8, 19), we then predict genetic changes of populations under future climate change scenarios.
We began by exposing a high-quality subset of 211 geo-referenced natural inbred A. thaliana accessions (18) to an experimental extreme drought event during the vegetative phase, which killed the plants before they could reproduce (Table S1). After two weeks of normal growth, plants were challenged by a terminal severe drought for over six weeks and imaged every 2-4 days (Fig. 1A) (see Supplementary Online Materials [SOM]). A polynomial linear mixed model was fit to the time-series data to quantify the rate of leaf decay (Fig. 1B-D, Video S1). The genotype deviations from the mean quadratic-term in the model provided the best estimate of this survivorship trait in late stages (Fig. S3, see details in SOM), ranging from −5 to +5 × 10−4 green pixels/day2. The most sensitive plants survived only about 32 days, while the most resilient plants survived about 15 days longer.
The amount of water available during the drought experiment translates to only about 30-40 mm of monthly rainfall, and as expected, accessions with higher survival come from regions with low precipitation during the warmest season (correlation with climate variable bio18 [www.worldclim.org. ref. (20)]: Pearson correlation, r=−0.19, p=0.005), and specifically with low precipitation during May and June (r≤−0.19, p≤0.005) (see Fig. 2A) (21). To further exploit current climatic data, we used 19 bioclimatic variables and random forest models (22) for environmental niche modeling (ENM) to predict the geographic distribution of the drought-survival index across Europe (Fig. 1C). Surprisingly, we found that individuals with higher drought survival were not only from the Mediterranean, but also from the opposite end of the species’ range in Sweden (Fig. 1C, ENM cross-validation accuracy=89%, Table S10) (21). In contrast to the warm-dry Mediterranean climate, Scandinavian dry periods occur on average at freezing temperatures (Fig. S12). Consequently, precipitation might occur as snow, which is not accessible for plants and produces a physiological drought response (23).
We then studied whether the different populations of A. thaliana are locally adapted (5) to low precipitation regimes via an increased drought-survival. Using an extended panel of 762 A. thaliana accessions (Table S1) we carried out genetic clustering (24) and studied population trajectories (25) (Fig. 2). This corroborated the existence of a so-called ‘relict’ group (12) and ten other derived groups of relict (e.g. Spanish groups) or other (e.g. Central Europe) origin; likely of the result of complex migration and admixture processes (26). A generalized linear model indicated that genetic group membership explains a significant amount of drought-survival variance (GLM: R2=12.8%; p= 4×10−5), with the North (N) Swedish and Northeastern (NE) Spanish groups each having on average higher survival than the other groups (t-test p≤0.01). A population graph estimated by Treemix (27) suggested a gene flow edge between the Mediterranean and Scandinavian drought-resistant genetic groups, potentially indicative of historical sharing of drought survival alleles (Fig. 2D). Finally, running an ENM of the genetic group membership with climatic variables from the origin of plants confirmed that the most important predictive variable is precipitation during the warmest quarter (bio18), followed by mean temperature of the driest quarter (bio9), and minimum temperature of the coldest month (bio6) (ENM accuracy > 95%. Fig. S8B and Table S10). As our results indicate that the deepest genetic structure parallels the local precipitation regimes and the ability of populations to survive drought, we expect that areas with the strongest decline in rainfall will see the most turnover in genetic diversity (see Fig.12 Fig. S8) (11).
Because the potential of populations to adapt to drought will depend on the genetic architecture of the selected trait, we identified drought-associated loci with EMMAX (28), a genome-wide association (GWA) method. Although genotype-associated variance (28) h2 was 50%, no individual SNP was significantly associated with drought survival (minimum p ~10−7, after FDR or Bonferroni corrections p > 0.05) (Fig. S5, Table S3). Significant associations in multiple phenotypes have been detected in similarly powered A. thaliana experiments (29). While multiple testing adjustment can over-correct p-values and obscure true associations, the absence of significant associations may also be due to (i) polygenic trait architecture, with many small-effect loci (30) and/or (ii) confounding by strong population structure, consistent with the association of drought survival with genetic group membership.
To test for polygenic adaptation, we repeated the GWA analyses with a model that specifically handles both oligo- and polygenic architectures, BSLMM (31). This model estimates, among other parameters, the probability that each SNP comes from a group of major-effect loci. Around half of the top non-significant EMMAX SNPs were found to have over 99% probability of belonging to such a major-effect group (Fisher’s exact test of overlap, p=3 × 10−7; see SOM). We further tested the polygenic hypothesis using the population genetic approach of Berg & Coop (32). The test is based on the principle that if populations diverge in drought-survival due to many loci, there should be an orchestrated shift in their allele frequency. After testing some 60 groups of EMMAX SNP hits of variable size and at different ranks, we detected the most significant signal of polygenic adaptation with the group that included the 151 top SNPs (Table S9). The signal was lost for ranks below the top 300-400 EMMAX SNPs (Table S9). We then compared summary statistics of the top 151 SNPs with background SNPs matched in frequency to avoid GWA discovery biases. The top 151 SNPs showed high Fst values, consistent with allele frequency differentiation between populations (Fig. S5). Tajima’s D values were positive (U Mann-Whitney p-value < 0.05), indicating intermediate allele frequencies at the GWA loci (Fig. S5), which could be a result of selection favoring alternative alleles in different ecological niches of the species (33). The top SNPs did not show any evidence for precipitous reductions of haplotypic diversity, as would be expected for hard selective sweeps (34) (Fig. S5). Together these patterns fit the expectations of local adaptation from a polygenic trait controlled by some hundreds loci (35) — theoretically expected to enable a fast response to a new environmental shift (36)
During local adaptation, the relevant loci diverge due to natural selection across populations, which generates a statistical correlation with population groups (37). In this situation, the default correction of population structure applied in GWA might obscure some of the true associations. While Fst scans can be useful to identify overly divergent loci across populations, elevated genome-wide Fst due to strong population structure can difficult outlier detection (37). as it is in our case (Fig. S4). In order to recover relevant variants that are deeply divergent across populations, we can study the ancestry of each SNP. Using ChomoPainter (38), which relies on linkage disequilibrium information, we segment each genome in question into its different population ancestries (here 11 groups). The first outcome of this analysis was that individuals from NW and NE Spain and, to lesser extent, the Southern Mediterranean (Fig. 2A), have inherited many DNA segments from relict individuals (Fig. S7). Then, in a generalized linear model framework, we test whether the ancestries of individuals at a SNP coincide with the observed phenotypic differences in drought-survival. Performing this “ancestry” genome-wide association (aGWA) and using a permutation correction of p-values (see SOM), we detected 8 distinct peaks (p<0.001, fig. 3A) including over 1,000 significant SNPs (70 SNPs after linkage disequilibrium pruning) (Table S4). The most prominent peak was located on chromosome 5 and explained over 20% of the variance in drought survival (Table S4). There was no overlap in top SNPs between GWA and aGWA because they search for different association signals. Our aGWA resembles other admixture mapping techniques (39). and might be most useful for associations in scenarios of adaptive introgression and local adaptation. To understand the origin of aGWA-identified SNPs, we constructed trees for all concatenated aGWA SNPs and for genome-wide background SNPs. Although the individuals from both the warm (Iberia and relicts) and cold (Scandinavia) edges of the species distribution are far apart in genome-wide SNPs, they are closely related in drought-associated SNPs (Fig. 3B). Overall, this is consistent with a common Mediterranean origin of drought-adaptive genetic variants of both Northern and Southern individuals (Fig. 2D, Fig. 3B), and highlights the relevance of populations at the latitudinal extremes of the species range as a possible genetic reservoir for future climate change adaptation (12).
Depending on the nature of the stress, different mechanisms for drought adaptation can be most advantageous (23, 40, 41). Annual plants, including A. thaliana, typically adapt to water stress deficit by accelerating the transition from germination to flowering (escape strategy) (14–16, 41) instead of increasing water use efficiency (avoidance strategy). Previous drought experiments with A. thaliana showed variation in both strategies but concluded it predominantly utilizes the drought escape strategy. Our extreme drought experiment focused in characterising the avoidance strategy by means of the drought-survival index, which was linearly associated to precipitation regimes (Fig. S11, Table S6). This trait was not correlated with flowering time of the accessions in unstressed conditions (Pearson correlation, r=0.07, p=0.12). However, we found a positive correlation between drought-survival and flowering time GWA summary statistics of the top 151 SNPs (Pearson correlation, r=0.51, p=1 × 10−11, see SOM) — suggesting a weak genetic trade-off (16). Interestingly, we did not find any associated between GWA or aGWA top SNPs and known flowering time QTLs (14–16). but rather a weak enrichment with membrane transporters (see SOM). Adjustment of osmotic balance through cell membrane transport is a drought avoidance mechanism (42) that might also confer cross-tolerance to other abiotic stresses (43), therefore it might be of relevance for Scandinavian A. thaliana accessions or other populations in the niche extremes (Fig. S12) (21).
Increased survival to extreme abiotic stresses should confer an evolutionary advantage given the predicted increase in drought frequency and intensity both around the Mediterranean and in Europe, which will constitute a critical hazard for many plants, including A. thaliana (2,11). Environmental niche models (ENM), which have been developed to relate species distributions to climate variables, can be used to predict future changes to species’ ranges (2, 3). Ignoring adaptation from standing variation (44–46), however, could lead to overestimates of extinction rates (47–49). By fitting ENM of current climate with SNP data (19). using a similar rationale as in Hancock and colleagues’ “climate GWA” (7), we can predict the most likely genetic makeup under current and future climate conditions. Using such an approach, we trained ENMs with 762 accessions and produced maps of the present distributions of the 151 GWA and 70 aGWA drought-associated SNPs (all ENM 5CV accuracy >92%; Table S3-4, Fig. S13–16). Concatenating the 221 maps, we inferred the most likely individual genotype at each location. At present, individuals from both Northern and Southern edges of the distribution are predicted to harbor more drought-survival alleles than those located in between (Fig. 3C, Fig. S15–16, with the quadratic term in a regression of allele count on latitude being positive at p=10−3), corroborating our previous observations. Then, using the trained ENM, we forecast the distribution of the 221 drought-survival alleles in 2070 (rpc 8.5, IPCC, www.ipcc.ch. ref. (20)). While it was expected that populations in the Mediterranean Basin would need to become more drought resistant (11), we predicted a more robust increase in the total number of drought-survival alleles for Central Europe (Fig. 3, Fig. S14–15). This is because rainfall in Central Europe will likely become more similar to that in the Mediterranean by 2070 (2, 11) (Fig. S12).
Because some of the drought-survival alleles are currently not yet present in Central Europe, we speculated that gene migration might be necessary to facilitate adaptation to future conditions (50), An underlying assumption of the ENM is that allele presence only depends on environmental variables, but this assumption, “universal migration”, may not be realistic for future predictions if present distributions are geographically narrow. We therefore included two geographic boundary conditions in the ENM to generate two models that were either more or less “migration-limited” (see SOM). After fitting all possible models and predicting allele distributions with future climate, we calculated the difference of predicted presence per map grid cell between the naïve, free migration ENM and the two geographically constrained ones (Fig. 3D-E). If an allele has currently a narrow distribution or is specific to a certain genetic background, its future presence in an area might not be predicted by the constrained models, even though the climate variables coincide with the SNP’s environmental range. Such a scenario seems to apply to Central Europe, as the deficit in drought-survival alleles predicted by the free over the constrained models was 8-30% (18-66 out of 221) (Fig. 3E; with the quadratic term in a regression of the allele count difference on latitude being negative at p < 10−10). Central European populations may therefore be under threat of lagging adaptation by the end of the 21st century.
In the end, for a population to persist, not only the number of drought-survival alleles has to increase, but it has to do so in actual individuals (51). The chance of this occurring will depend on local allele frequencies and the natural selection favouring the drought-survival alleles. Therefore, we studied current allele frequencies at three representative locations with the highest sampling density in our dataset (40 samples within a 50 kilometer area): Madrid (Spain), Tübingen (Germany) and Malmö (Sweden), which are at the southern edge, center and northern edge of the range, respectively. Based on ENM predictions, we calculated allele frequencies from present to 2070. Frequencies are predicted to increase significantly only in the Tübingen population (Student’s t test, p<10−16, Table S11), but not in Madrid and Malmö, indicating that these two populations might be already adapted to the future local climate. Because the Tübingen population already has most drought-associated alleles (53% of 70 aGWA SNPs and 90% of 151 GWA SNPs), increasing the number of total favorable alleles in individual genotypes should be feasible, especially since there are single genotypes that have 63% (aGWA) and 90% (GWA) of those alleles already present (see SOM). Starting 50-generations simulations at the present Tübingen frequency of independent drought-survival alleles and assuming a range of selection coefficients, we estimated that a 1-3% of fitness advantage on average would be necessary to increase frequencies to match those of the adapted Madrid and Malmö populations (Fig. S17, see SOM). Such selection could take place efficiently in large populations like the ones of a highly-reproductive weed (51, 52).
Leveraging the model organism A. thaliana, we have begun to address key questions to understand the burning issue of climate change effects on biodiversity. We provide evidence for the possibility of adaptive genetic variation to extreme drought events. Harnessing the power of methods that allow polygenic genetic architecture and testing evolutionary hypotheses of natural selection, we detected that relevant genetic variants had been under polygenic local adaptation and were more abundant at the edges of the species range. Extreme adaptation at range edges might indeed be critical for a species’ persistence under climate change. Although many aspects of future adaptation are not considered here, namely non-drought related or seasonal climate change (51). biotic interactions, phenotypic plasticity, or novel adaptive mutations (53), our spatially explicit analyses emphasize the potential of adaptive evolution from standing variation to ameliorate climate change’s detrimental effects.
Author contribution
MEA conceived and designed the project. GW and FV helped and advised on image phenotyping and FV provided additional phenotypes. MEA and WD performed chromosome painter analyses. MEA performed the drought experiment, processed the image data, and designed and carried out the statistical analyses. DW and HAB advised and oversaw the project. MEA wrote the first draft and together with HAB and DW wrote the final manuscript with input from all authors.
Acknowledgements
We thank I. Henderson for the recombination map, R. Wedegärtner for assistance with the greenhouse drought experiment, the Petrov, Coop, Ross-Ibarra, Gaut and Schmitt labs for discussions. We thank J. Lasky, X. Picó, A. Hancock, H. Thomassen, T. Mitchell-Olds, J. Mujica, P. Lang, and D. Seymour for comments and the Weigel and Burbano labs for discussion. This work was supported by ERC Advanced Grant IMMUNEMESIS to DW and the President’s Fund of the Max Planck Society, project “Darwin” to HAB.