Progress report for GNE20-237
Seedling diseases are one of the top five most destructive soybean diseases in the Northern United States with an estimated 500,000 bushels lost in Pennsylvania in 2018 (Crop 2020). Pythium spp. is one important soilborne pathogen that can be suppressed by soil organisms like Trichoderma or increase in severity in the presence of other soilborne pathogens. This study’s objective is to further elucidate the microbiome’s role in Pythium disease development by assessing how microbiome composition shifts in bacterial and fungal abundance and diversity during Pythium disease development is predictive of disease severity in soybean. This interaction will be assessed under common management practices and soil conditions including fungicide seed treatments, liming, flooding, and soil sand/silt/clay content using a novel rhizobox design for non-destructive daily rhizosphere sampling in greenhouse and field experiments. Pythium qPCR, 16S-rRNA bacterial gene sequencing and ITS fungal gene sequencing using MiSeq will be used to visualize the abundances and diversity of fungal and bacterial microbiome members.
Mixed analyses including principal coordinates analysis, linear discriminant analysis, and correlation network analysis will identify treatment variations and relate microbiomes to disease severity. I hypothesize that certain management practices support a suppressive microbiome and that potential biocontrol organisms will be identified that can support sustainable agriculture. Results will be published in a peer-reviewed scientific journal, presented at professional and farmer meetings, and compiled with videos and photos from the experiments for public dissemination as factsheets, online educational resources, and extension teaching materials to promote adoption of sustainable management practices.
Objective 1: Determine how common PA soybean production management practices and field conditions such as fungicide seed treatments, liming, flooding, and soil sand/silt/clay content impact shifts in bacterial and fungal composition and diversity during pre- and post-emergence Pythium disease in soybeans up to two weeks after planting using both greenhouse and PA research field experiments.
Objective 2: Determine how shifts in soil microbiome fungal and bacterial diversity and abundance during Pythium disease development relate to the severity of disease after two weeks from planting.
Objective 3: Identify key fungal and bacterial taxa in the soil microbiome involved in synergism with or inhibition of Pythium disease development under the different management and soil conditions, supporting future studies delving into the pathogen-interaction mechanisms that can be developed into viable sustainable management options such as biocontrols.
The purpose of this project is to identify keystone species or microbial groups in the soil microbiome associated with increased Pythium disease severity under different soil conditions and management practices to determine the indirect effects of these practices on disease outcomes. This is approached by determining shifts in bacterial and fungal soil microbiome abundance and diversity during Pythium disease development in soybean grown during the initial two-week emergence period after planting. Results will inform the timing and choice of sustainable management tactics that reduce the impact of Pythium and other early seedling diseases.
Seedling diseases are regularly one of the top five most destructive soybean diseases in the Northern US and Ontario, Canada, with over 60 million bushels lost to these diseases in 2014 (Allen et al 2017). Pythium is a ubiquitous soilborne oomycete and one of several pathogens causing pre- and post-emergence damping off, or seedling death from root rot, in multiple economically important crops like corn and soybean. It is especially problematic in the Northern US due to the wet weather and cooler conditions at planting that promote disease development. Disease severity is also influenced by other soil organisms, such as suppression by Trichoderma or increased severity with Fusarium. Pythium is commonly treated using fungicide seed coatings; however, these do not always perform well and the reason is not completely clear. I hypothesize that the fungicide’s unintended impacts on the microbiome and natural beneficial soil organisms plays a role in why they fail; therefore, I will test these treatments as well as other soil and management conditions in this project. My research will improve knowledge on the unintended effects of fungicides and promote consistent efficacy for long-term use.
Overall, this project contributes to agricultural sustainability by identifying best practices promoting a disease suppressive soil microbiome that can protect crops across growing seasons. I hypothesize that we will identify future biocontrol organisms that can be introduced into soils for long-term benefits and reduce farmer’s dependence on fungicide seed treatments. As soybean prices fluctuate from trade wars, weather losses, and changing demand the extra cost of approximately $10/acre for fungicide treatments are risky given their variable efficacy against Pythium and other soilborne organisms. Additionally, Pythium populations have been known to develop fungicide resistance, such as to metalaxyl, and rotating management strategies can decrease the risk of this occurring. It is important for the economic sustainability of the farm enterprise to have reliable alternatives for treating seedling diseases that reduce inputs while supporting productivity. Additionally, public health and environmental concerns and costs associated with fungicide use can be alleviated. Manipulating the soil microbiome is an environmentally conservative way to address soilborne seedling diseases like Pythium for the long term. Results from these studies will be disseminated to different stakeholders to improve understanding of the role of the microbiome in seedling disease development, I will create educational resources for farmers, extension, and scientists for use in presentations, to reference for management decisions, and for developing new research projects surrounding agricultural sustainability.
I hypothesize that under different management or soil conditions different fungal and bacterial microbiome compositions will be present and that throughout Pythium disease development, these compositions will change in both abundance and diversity. I expect to see commonalities among several key taxa across treatments that indicate a relationship with Pythium disease or host plant development. To test this, I will conduct a greenhouse experiment to isolate individual plant associated soil microbiomes during Pythium disease development, followed by a research field experiment to verify these interactions when plants exist in communities and soil microbes can migrate more freely. Soil for the greenhouse will be sourced from a primary research field site with a history of soybean production (Penn State Rock Springs Research Farm located close to the University Park campus) and amended for the treatments following baseline soil testing at the Penn State Soils Lab. Soil will be collected from the top 10 inches of soil and put directly into bleach sterilized rhizoboxes. For control purposes, all soils will be inoculated with a pathogenic Pythium spp. isolated previously from a PA soybean field to ensure pathogen presence. A single soybean variety with untreated and fungicide treated seeds will be grown in rhizoboxes. Plants will be grown for two weeks from planting in both the greenhouse and field.
There will be four different management or soil conditions that are present in PA soybean production systems considered: soil types, pH levels, soil saturation, and fungicide seed treatments. Each treatment will have three replicates to verify that shifts seen are statistically similar across plants in each treatment. Soils are unsterilized due to the nature of the study to preserve the inherent microbiome as closely as possible.
- For soil type comparison, the sandy loam soil from Rock Springs, representing Eastern PA’s Ridge and Valley Soils Region, will be paired against a silty clay soil from the Western PA Pittsburg Plateau Soils Region obtained through my connections with the PA Soybean On-Farm Network. The Pittsburg Plateau soil will also undergo a soils test so that baseline soil characteristics can be compared.
- Considering that PA soils are naturally acidic, I will use lime amendments to assess high (~8) and low pH (~5.5) scenarios which represent the upper and lower limits of Pythium disease development from the literature (Kauraw 1979). The Rock Springs soil from the soil type treatment will provide a mid-range pH level for comparison.
- The seed fungicide ApronMaxx, used in PA Soybean On-Farm Network trials, will be used to determine how fungicides impact non-target organisms in the soil microbiome that play a role in Pythium disease development and allows us to strengthen our recommendations for seed treatment fungicide use.
- Climate change has increased precipitation in PA and saturated soil is both a plant stressor and a key abiotic driver behind Pythium disease development. It also plays a role in the fate of fungicides and their efficacy; therefore, saturated soil with and without treated seeds will be considered.
There will be six experimental controls plants grown in sterilized Rock Springs soil, with half inoculated with Pythium. Plants grown without Pythium provide a baseline for plant growth in a disease-free environment, while inoculated controls offer a comparison for disease severity and microbiome changes given no management practices. A proxy control will be unsterilized Rock Springs soil used in the soils type test; therefore, five rhizoboxes will be used for that treatment. This soil can provide an additional treatment for comparison in the different management scenarios, particularly acting as the untreated seed treatment in the fungicide assessment.
Greenhouse vs Field Experiment Design
The setup of the field and greenhouse portions of this study are slightly different to balance sufficient data for robust data analysis with having manageable numbers of samples for sequencing. The greenhouse experiment, conducted first, includes all treatment and control replicates described above for a total of 29 plants. Greenhouse conditions, such as relative humidity, temperature, sunlight, and plant care including watering will be uniform across rhizoboxes, with the exception of extra watering for soil saturation treatments. Plants will be randomly assigned to a new location in the greenhouse each day to avoid location bias.
After greenhouse plants are harvested, leftover soil will be discarded and rhizoboxes will be bleach sterilized prior to commencing with the field experiment. New Rock Springs soil will be used to avoid carryover effects of the disease and microbiome shifts from the greenhouse study. Field plots will be amended with the respective treatments and first planted with soybean seeds before a slice of soil is removed for the rhizobox set with a single seed. Rhizoboxes will return to the same location each day after sampling. This simulates field conditions where plants grow in proximity to each other and microbes can move from the surrounding soil in and out of the rhizosphere. There is minimal control over environmental factors, although flooded treatments will receive water to maintain saturation.
There will be no sterilized controls for the field study; unaltered Rock Springs soil will act as the control with six replicates of which three will be inoculated with Pythium for its use as a comparison to the different treatments. There is no soil type treatment proposed for the field experiment. However, we can use the PA Soybean OFN as a surrogate to collect soil samples from different farms to conduct sampling at pre-planting, 7 and 14-days after emergence to compare the role of soil type with greenhouse results. The soil type OFN validation work is not part of this proposal, as it would be covered under OFN grant funding. All other parts of the experimental design and data collection process will remain the same as in the greenhouse. Altogether, 21 plants will be grown in rhizoboxes for the field experiment.
The novel use of rhizoboxes has the advantage of enabling multi-day non-destructive rhizosphere soil sample collection and use over multiple experimental runs with sterilization. Initial tests of 3D printed rhizobox prototypes for use in this experiment, designed with Tinkercad, have shown promise and these would allow for greater uniformity of the boxes and provide a template that other researchers can use in future experiments. These opaque square boxes have holes in the bottom for drainage, holes for wire handles at the top (to make it easier for daily removal in the field), and holes in the sides for sampling. The sampling techniques being evaluated for this experiment include inserting a small probe into all four sides to obtain a daily representative rhizosphere sample for analysis or the placement of soil-filled straws through the boxes located next to the seed (and directly below where the taproot will grow) that can simply be removed and stored at the time of sampling. Holes will also allow for microbial movement between the rhizobox and the surrounding soil in the field setting. The cost of 3D printing these boxes may allow for additional plants to be grown for the experiment that I can collect and store extra samples from.
Daily sampling of plants will occur over two weeks in order to capture Pythium disease at (1) soybean pre-emergence and (2) during emergence. This allows for determining the relative rate of soil microbiome changes and best sampling frequency that can be used for future studies. A sample will be taken prior to planting to identify the original microbiome composition, then several hours post-planting when Pythium can start attacking seeds. Subsequent sampling will occur once per day until day 14 when seedlings are harvested for visual disease assessment.
Sequencing & Analysis
Initial sequencing will be done for each sampling day for one random replicate from each treatment. The other two replicates will have the following samples initially sequenced: pre-planting, day one-post planting, then (1) sequencing on days 3, 5, 7, 9, 11, 13, and 14 (harvest) while (2) is sequenced days 2, 4, 6, 8, 10, 12, and 14. This results in a manageable number of samples for sequencing while still ensuring enough samples for a robust statistical analysis. Soil samples from unsequenced days will be stored for future analysis as needed for statistical purposes. Overall, 15 samples are collected from each plant and a subset of these will be initially analyzed.
DNA extraction for each soil sample will be used to run three PCR sets including 16S-rRNA bacterial gene sequencing, ITS fungal gene sequencing and qPCR for Pythium to verify and quantify its presence in the microbiome. 16S-rRNA and ITS PCR products will be used for MiSeq sequencing at the Penn State Genomics Core Facility. Sequencing data will be processed using UPARSE and UCHIME in the open-source software Mothur.
Principal coordinates analysis in R will be used to visualize soil microbiome compositions at each timepoint, with variation in initial soil physicochemical properties visualized using principal components analysis. AMOVA and significance tests will be used to understand differences between treatments. Tax4Fun, also run in R, will be used to predict functional capabilities of the microbiome from the 16S sequencing data. Altogether, this genomics data will be used to visualize the abundances and diversity of fungal and bacterial microbiome community members throughout Pythium disease development.
In this objective, my hypothesis is that abundance of particular fungal and bacterial taxa in the soil microbiome at different timepoints in Pythium disease development will be predictive of Pythium disease severity. To test this, Pythium will be first classified as pre- or post-emergence disease. Photos will be taken at each sampling timepoint to monitor disease development visually. After 14 days plants will be harvested, photographed, root length and spread measured (centimeters), and shoot and root dry weight biomass (grams) determined to relate to disease severity.
Root symptoms for visual ratings include lesions, brown discoloration, and rot, while seedling symptoms focus on water-soaked leaf and stem lesions, blue-green to brown discoloration, ‘mushy’ wilted seedlings and death. Root severity ratings derived from the literature will be assessed for all plants on a scale of 0 to 9, where 0 = <10% of root exhibits symptoms, 1 = 10-20% of root exhibits symptoms, etc., 9 = plant dead or germination failure. Post-emergence diseased plants will be rated on a 5-point scale of presence of damping-off symptoms where 0 = no visual disease symptoms (healthy plant), 1 = few small lesions or light discoloration, 2 = mid-sized or many small lesions, moderate discoloration, 3 = large or many midsized lesions, severe discoloration and wilting occurring, 4 = completely wilted and mushy (plant dead). Seedling symptom ratings will be linked to their root rot severity in analysis.
Building off the statistical analysis under Objective 1, linear discriminant analysis and significance tests will be used to identify key taxa and their relative abundances at different timepoints related to final Pythium disease severity.
The hypothesis for this objective is that several key fungal and bacterial taxa in the soil microbiome will drive severity or suppression of Pythium disease. To test this, I will use results from Objective 1 to conduct a correlation and co-occurrence network analysis using a combination of the open-source software R, Netshift, and Cytoscape. These analyses identify highly interactive “hub” microbes and determine their potential for being keystone microbiome members during Pythium disease development by quantifying directional changes in individual taxa interactions. Results from this objective are important to provide insight into organisms that can be considered as future biocontrol options or alternative organisms to target with management practices to suppress Pythium disease severity.
This research is still in its initial pre-experiment phase. Experiments will be conducted in April & May 2021, as initially outlined in the project timeline.
This research is still in its initial pre-experiment phase. Experiments will be conducted in April & May 2021, as initially outlined in the project timeline.
Education & Outreach Activities and Participation Summary
A major goal of this project is to provide the scientific community fundamental knowledge of how management practices impact the soil microbiome compositional shifts during Pythium disease development and to identify keystone taxa with potential to either suppress or increase disease severity. These results are important to support in-depth studies of these alternative microbe-pathogen interactions and biocontrol mechanisms while also promoting sustainable management production practices. Most educational and outreach aspects of this project will be completed in 2022, as outlined in the original project timeline.
I will establish a dedicated project website in February 2021 that is updated throughout the study with basic information regarding the project, its progress and final results, PI contact information, and related resources that people can use for educational or research purposes. Project information and results distribution will also occur via a Penn State News article at the end of the project, which is regularly received by all members of the university. Results from this research will be submitted to a peer-reviewed journal for publication and published as an abstract to present at the 2022 Annual Meeting of the American Phytopathological Society (APS). I will also present this research at the Graduate Student Exhibition on the Penn State Campus in Spring 2022 and as part of the Penn State Department of Plant Pathology and Environmental Microbiology Departmental Lecture Series in April 2021 and Spring 2022, which is open to the public and other researchers in-person and via Zoom.
Photos of the treatment replicates and experimental design will be compiled to illustrate the methodology to support future research and develop educational materials on Pythium disease development stages and severity. Additionally, videos will be taken of various experimental processes such as setting up the experimental design and sample processing for sequencing to teach students about important lab protocols and good practices, researchers on new methodologies and experimental design, and to educate the farmers and public who are typically disconnected from the off-farm side of research on scientific methods. Adobe products such as InDesign, Illustrator, & Premiere Pro will be used for editing and compilation into polished videos, factsheets, and pamphlets. Most outreach materials will be developed in 2022, after the field and greenhouse experiments are completed and data is analyzed.
These resources will be available on the project website and will be actively shared through extension, social media platforms such as Facebook and YouTube, and scientific communities such as APS. Results and educational materials from this project will be presented to soybean farmers through the PA Soybean OFN, a program that annually has four summer field days and four winter workshops. A free peer-reviewed publication in the APS online journal, The Plant Health Instructor, will focus on an educational overview of how the soil microbiome supports or suppresses Pythium disease given different management and soil conditions. I will also publish educational articles in Field Crop News from Penn State Extension and the Field and Forage Crops Group. Most publications will be submitted in 2022, after the field and greenhouse experiments are completed and data is analyzed.
This project is in its initial pre-experiment phases; this section will be updated as the project progresses and data is analyzed. I expect that the knowledge generated from this project will help farmers make educated management decisions such as related to reducing fungicide usage which has economic, environmental, and social benefits. Additionally, I expect to identify potential biocontrol candidates and characterize a Pythium suppressive soil microbiome that can limit disease in soybean fields long-term with limited inputs.
This project is in its initial pre-experiment phases; this section will be updated as the project progresses and after the experiments occur. At this stage I would simply say that research to support sustainable agriculture requires a multi- and inter-disciplinary approach. My mentor is an engineer and I never would have thought to try 3D printed boxes for the experiment if I hadn’t discussed it with someone outside of my discipline. I am very interested in pursuing interdisciplinary research in my future career because I enjoy the refreshing experience of seeing the same research project from different perspectives and integrating those perspectives to develop and conduct successful research endeavors.
This project is in its initial pre-experiment phases; this section will be updated as the project progresses.