Dr. Jordan Dowell (Louisiana State University) studies the evolutionary ecology of plant-plant chemical communication and the impacts of multifunctional traits on biotic interactions from single-cells to landscapes using various techniques, from metabolomic and genomic approaches to remote sensing and field-based studies.

Research Interests

As a chemical ecologist and comparative biochemist, I am especially interested in understanding the evolution of chemical diversity and the effects of chemical diversity on organismal interactions across spatial and temporal scales. As nature’s prime organic chemists, plants produce diverse chemicals (metabolites) to deter and attract beneficial organisms, ease abiotic stress, and communicate within and among species. Producing metabolites to mediate biotic and abiotic interactions requires nutrient investment, often at the expense of growth or reproduction. For instance, photosynthesis, specifically the assimilation of CO2 into organic molecules, is a fundamental ecosystem process. Assimilated carbon can be invested into more nutrient-acquiring tissues to increase photosynthesis(e.g., growth) or reproduction to promote fitness.

However, plants live in dynamic, stressful environments. As a result, plants face an inherent physiological trade-off between investment in specialized metabolites that mediate biotic and abiotic stress versus investment in growth or reproduction to ensure individual or population survival. However, ‘tradeoffs’ suggest that investment in specialized metabolites has no role in growth or reproduction, but as with most biological concepts, ‘it depends.’ For instance, volatile organic compounds (VOCs) attract and direct pollinators, increasing fitness, and can communicate herbivore damage within and among individuals to prime induced defense pathways. In trying to go from ‘it depends’ to ‘under these set of circumstances”, my program asks, “How do multifunctional metabolites fit into evolutionary and ecological hypotheses,” and more simply, “How many metabolites are multifunctional !?”

Traditional cell-signaling and chemical ecology hypotheses assert that single molecules are responsible for these induced responses. However, in VOC-induced responses in plants and fungi, qualitative and quantitative variation in VOCs blends have a larger impact than any single VOC in inducing plant responses, suggesting signal synergism. As VOCs are structurally constrained, synergism may have developed due to similarity in physiochemical properties. Further, maintaining synergistic signaling may allow for metered responses using conserved protein receptors, while shifting cues may help conceal plants from organisms with more canalized sensory perception, like animals. Thus, selection by pests and pathogens may drive changes in cues more directly than cellular receptors, leading to increased diversity of signaling molecules compared to receptor proteins.

In addition to signaling, VOCs act directly and indirectly. For instance, cis-3-Hexen-1-ol (the smell of fresh-cut grass) is widely distributed across the plant kingdom and can induce acquired pest and pathogen resistance across many plant species. In addition, this VOC directly inhibits the sporulation and growth of a common host-generalist fungal pathogen, Botrytis cinerea. However, in the broader scale of host-pathogen interactions, there are no definitive studies or models delineating the impacts of maintaining detoxification mechanisms and, simultaneously, retaining the ability to interpret host-infochemicals on pathogen phenotypic and genetic diversity.

To answer these questions and ultimately ask more, my group combines integrative, multi-omic approaches with natural and life history to take an organism-centered approach. My program has three primary foci: (1)Developing theoretical and empirical models of the evolutionary ecology of plant VOC-mediated interactions, (2) Identifying the relationships between the evolution of phytochemical diversity and physiological tradeoffs, (3) Testing the direct and indirect effects of plant VOC-mediated interactions in plant-pathogen systems, and (4) Developing statistical frameworks to assess chemical diversity and models of chemically-mediated cellular and species-level interactions.

Specific Foci

type stuff

Focus 1

VOC-induced responses among individuals within a species have reduced effectiveness as populations’ genetic and physical distance increase, but we observe a high efficacy of VOC-induced responses among distantly related species. We identify biotic and abiotic factors contributing to shifts in the efficacy of VOC-induced responses across evolutionary scales. my group hypothesizes that selection and genetic drift impact the transition from plant ‘communication’ as a function of self-recognition towards ‘eavesdropping, manipulation, or mutually beneficial communication’ mediated by ecological interactions, including physiological tradeoffs, niche partitioning, biotic stress intensity, and reproductive viability. To test these hypotheses, we conduct cross-species experiments in lab and field settings using multi-omic approaches using the annual clade of Helianthus (Sunflower) as a model system. The sunflower system makes a great model as many species vary in hybridization success and ecological niche, being native to all ecoregions of the continental U.S. So far we have assessed physiological variation, constitutive and herbivory-induced VOCs, and developed a noninvasive tool to assess induced responses across the genus in real-time. From these experiments, the explanatory power of factors affecting VOC-induced responses expands our understanding of the interaction between community assembly and evolutionary processes. Further, including transcriptomics to understand the regulation of VOCs and identify plant receptors (none of which have been identified or hypothesized!) will lead to novel avenues of signaling research.

Focus 2

Developing evolutionary models requires understanding the regulation of trait variation and inheritance. My lab tests how offspring from chemotypically and physiologically distinct parents vary compared to transcriptomic and genomic predictions. In addition, we examine how hybridization and introgression alter patterns of metabolite profiles and physiological variation with experimental crosses. We use the comprehensive examination of the inheritance of phenotypic variation across the Helianthus genus to develop novel hypotheses to test more broadly in other species of economic and ecological importance. In our previous genome-wide association studies(GWAS) in cultivated Sunflowers, we have found that lines selected for increases in absolute fitness (e.g., total seed yield) are associated with leaves with high photosynthetic capacity and reduced concentration of specialized metabolites in comparison to lines selected for fitness stability (stress tolerance) associated with leaves with lower photosynthetic capacity and higher concentrations of specialized metabolites. In addition, several linkage blocks were associated with both physiology and specialized metabolism. These data point towards pleiotropy or selective sweeps underlying observed phenotypic variation.

Focus 3

Assessing direct and indirect impacts on fitness requires delineating effects across generations. We are conducting a multi-generational toxicological GWAS of VOC effects on fitness-related traits in Botrytis cinerea. In addition, we are testing the impacts of VOC exposure(to host and/or pathogen) on the infection success of new tissue. Using and expanding this data, my lab tests how multifunctional metabolites, mixture toxicity, and host-induced responses can affect pathogen genetic diversity and life history evolution. As my lab continues, simulations will be combined with further data on isolate-level competition, nutritional dynamics, and microbiome composition to produce and test novel hypotheses about the evolutionary ecology of host-pathogen interactions in the context of multifunctional host-infochemicals (VOCs).

Focus 4

As we continue to develop new questions concerning the evolutionary ecology of complex traits, our methods also need to develop. Like individual species, metabolites are non-independent entities. Thus in examining metabolomic data, we need to account for variation due to shared biosynthetic properties accurately, as well we can exploit autocorrelation for synthetic biology. To understand the explanatory power of canonical and non-canonical biosynthesis, we developed a statistical method to assess the phenotypic integration (complex patterns of covariation among functionally related traits) of metabolite profiles, given an unknown pathway using the empirical three-dimensional physiochemical properties of observed metabolites. To integrate our previous work into synthetic approaches to assess physiological tradeoffs, we am incorporating process-based models of metabolism to assess energetic investment in various pathways and the integration of metabolism. We are examining how pathogen infections alter carbon and nutrient investment by Arabidopsis thaliana into specialized metabolism. In tandem, we are modeling investment by the fungal pathogen B. cinerea into specialized metabolism during infections. By incorporating transcriptomic data, we are developing proxy measures of metabolism and identifying reactions associated with susceptibility or virulence. Specifically, we use these models to explore the evolution of metabolic flux and energetic investment in specialized metabolic products based on simulated nutrient requirements necessary for production, leveraging publicly available transcriptomic datasets. In addition, as single-cell transcriptomic data becomes more available, we will apply these pipelines to assess cellular physiological tradeoffs and explore how the metabolism of organs is coordinated. Further, as most studies examine physiological tradeoffs in the context of carbon, nitrogen, or water acquisition, My lab will expand the field by exploring alternative physiological tradeoffs in the context of various macro and micronutrients through simulation and follow-up empirical validation