The "mustard oil bomb" is a major defense mechanism in the Brassicaceae, which includes crops such as canola and the model plant Arabidopsis thaliana. These plants produce and store blends of amino acid-derived secondary metabolites called glucosinolates. Upon tissue rupture by natural enemies, the myrosinase enzyme hydrolyses glucosinolates, releasing defense molecules. Brassicaceae display extensive variation in the mixture of glucosinolates that they produce. To investigate the genetics underlying natural variation in glucosinolate profiles, we conducted a large genomewide association study of 22 methionine-derived glucosinolates using A. thaliana accessions from across Europe. We found that 36% of among accession variation in overall glucosinolate profile was explained by genetic differentiation at only three known loci from the glucosinolate pathway. Glucosinolate-related SNPs were up to 490-fold enriched in the extreme tail of the genome-wide F ST scan, indicating strong selection on loci controlling this pathway. Glucosinolate profiles displayed a striking longitudinal gradient with alkenyl and hydroxyalkenyl glucosinolates enriched in the West. We detected a significant contribution of glucosinolate loci toward general herbivore resistance and lifetime fitness in common garden experiments conducted in France, where accessions are enriched in hydroxyalkenyls. In addition to demonstrating the adaptive value of glucosinolate profile variation, we also detected long-distance linkage disequilibrium at two underlying loci, GS-OH and GS-ELONG. Locally cooccurring alleles at these loci display epistatic effects on herbivore resistance and fitness in ecologically realistic conditions. Together, our results suggest that natural selection has favored a locally adaptive configuration of physically unlinked loci in Western Europe.Arabidopsis thaliana | glucosinolates | genome-wide association mapping | linkage disequilibrium | adaptation

Although the complex interactions between hosts and microbial associates are increasingly well documented, we still know little about how and why hosts shape microbial communities in nature. We characterized the leaf microbiota within 200 clonal accessions in eight field experiments and detected effects of both local environment and host genotype on community structure. Within environments, hosts’ genetics preferentially associate with a core of ubiquitous microbial hubs that, in turn, structure the community. These microbial hubs correlate with host performance, and a GWAS revealed strong candidate genes for the host factors impacting heritable hubs. Our results reveal how selection may act to enhance fitness through microbial associations and bolster the possibility of enhancing crop performance through these host factors.TextHosts harbor complex microbial communities that are thought to impact health and development1. This is best studied in human hosts for which the microbiota has been implicated in a variety of diseases including obesity and cancer 2. Efforts are thus underway to determine the host factors shaping these resident populations 3,4 and to use next-generation probiotics to inhibit colonization by pathogens 5. Similarly, in agriculture, there is great hope of shaping the composition of the microbiota in order to mitigate disease and increase crop yield in a sustainable fashion. Indeed, the Food and Agriculture Organization of the United Nations has made the use of biological control and growth promoting microbial associations a clear priority for improving food production 6.Plant associated microbes can be beneficial in many ways including improving access to nutrients, activating or priming the immune system, and competing with pathogens. For example, seeds inoculated with a combination of naturally occurring microbes were recently found to be protected from a sudden-wilt disease that emerged after continuous cropping 7. Thus, it would be advantageous to breed crops that promote the growth of beneficial microbes under a variety of field conditions, a prospect that is made more likely by the demonstration of host genotypic effects on their microbiota 8–10. That said, microbial communities are complex entities that are influenced by the combined impact of host factors, environment and microbe-microbe interactions 11. As a consequence, the extent to which host plants can control microbial communities to their advantage, especially in a natural context, is unclear.Here, we combine large scale field experiments of plant genotypes grown in their natural environments, extensive microbial community analysis, and genome-wide association mapping to (i) disentangle how the influence of the host is distributed among microbial community members, and thus how host variation shapes the microbiota, (ii) propose plant genes and functions that correlate with variation in the microbiota across environmental conditions, and (iii) examine how key microbial associates impact plant fitness. Our motivation is to further the goal of generating plants with an enhanced ability to host beneficial microbial communities.

Pollen-limited plants are confronted with a difficult tradeoff because they must present showy floral displays to attract pollinators and yet must also minimize their apparency to herbivores. In these systems, traits that increase pollinator visitation may also increase herbivore oviposition and overall plant resistance may therefore be constrained to evolve largely as a correlated response to selection on plant apparency or vigor. We used a familystructured quantitative genetic experiment to evaluate the importance of ungulate browsing, flowering date and plant height (traits that are related to overall vigor), and variation in a putative phytochemical defense (cucurbitacin production) on patterns of seed fly attack in a scarlet gilia population. We found significant genetic variation in the amount of insect damage plants experience in the field, providing evidence that resistance may evolve. In addition, we found that browsing reduced seed fly attack and that oviposition is strongly related to plant size and flowering date; large, early flowering plants experience high attack. In addition, we found that high cucurbitacin production was correlated with low seed fly damage, although this effect was relatively weak.We found directional selection on final plant height and flowering date; tall, early flowering plants had the highest reproductive success. In addition, we found negative directional selection on cucurbitacin production, which may indicate a high cost of cucurbitacin or other functions of this phytochemical. Although seed fly herbivory arguably decreases plant fitness, we found an unexpected positive relationship between damage and fitness. A negative relationship between fitness and damage may be masked in this system through strong positive indirect correlations between patterns of damage and levels of pollinator visitation. Finally, we found significant genetic variation in flowering date, plant height, and cucurbitacin production. Resistance to seed flies may evolve in this population, but largely as a non-adaptive correlated response to selection on overall plant vigor. Phytochemicals may play a more important role in defense in years with high seed fly attack, or when pollen-limitation is less severe.

Resource patchiness influences consumer foraging, movement, and physiology. Fluxes across ecosystem boundaries can extend these effects to otherwise distinct food webs. Intraspecific diversity of these cross-ecosystem subsidies can have large consequences for recipient systems. Here, we show intraspecific variation in leaf defensive chemistry of riparian trees drives local adaptation among terrestrial and riverine decomposers that consume shed leaf litter. We found extensive geographic structuring of ellagitannins, diarylheptanoids, and flavonoids in red alder trees. Ellagitannins, particularly those with strong oxidative activity, drive aquatic leaf decomposition. Further, spatial variation in these leaf components drives local ecological matching: in experiments using artificial food sources distinguished only by the chemical content of individual trees, we found decomposers both on land and in rivers more quickly consumed locally derived food sources. These results illustrate that terrestrial processes can change the chemistry of cross-ecosystem subsidies in ways that ultimately alter ecosystem function in donor and recipient systems.

Cucumber seeds were germinated under various combinations of solution volume and seed number with a range of ferulic acid concentrations. At each concentration, radicle growth decreased as the relative amount of ferulic acid available per seed increased from χ (25 seeds/5 ml) to 5χ (5 seeds/5 ml) to 19χ (25 seeds/95 ml). With 2.0 mM ferulic acid in buffered solution, radicle lengths after 48 hr ranged from 71 to 47% of control. The amount of ferulic acid remaining in 2.0 mM solution after 48 hr was directly proportional to the amount initially available per seed, and ranged from 9 to 91%. Solution volume and seed number also significantly affected inhibition by vanillic acid, caffeic acid, and juglone. With 0.1 mM juglone, radicle lengths after 48 hr were 88% of control with 25 seeds/5 ml, 68% with 5 seeds/5 ml, and 56% with 25 seeds/90 ml. The data demonstrated that lower phytotoxin concentrations can produce equivalent or greater inhibitory effects than higher concentrations when the amount available per seed for uptake is greater. Equivalent inhibition of radicle growth was observed with 1.0 mM (5 seeds/5 ml) and 2.0 mM (25 seeds/5 ml) ferulic acid. Available literature on herbicides indicates that similar effects occur in greenhouse and field studies.