Performing Department
(N/A)
Non Technical Summary
Nearly $17 billion was spent worldwide on fungicides in 2020, however the impacts of fungicides on plant microbiomes remains poorly understood. The authors recently collected preliminary data in turfgrass that showed a 6-fold increase in disease following the application of the broad-spectrum fungicide chlorothalonil relative to a non-treated control. The preliminary study was unable to determine the molecular mechanisms of this 'disease resurgence', and in this proposal we seek to employ a more robust microbial community and metabolomic analysis to unravel the mechanisms behind the dysbiosis that occurs to plant microbiomes and microbial processes following fungicide applications in both turfgrass and corn. We hypothesize that certain, broad-spectrum fungicides disrupt the resident plant microbial community and decrease the ability of the community to suppress pathogen activity once the pathogen-suppressing activity of the fungicide has dissipated. The information gained in this project will be used to identify fungicide chemistries that cause disease resurgence in turfgrass and corn, understand key microbial taxa and microbial processes disrupted by chemical fungicides, and lead to the development of more effective biocontrol strategies that reduce reliance on synthetic fungicides in a broad range of agronomic and horticultural systems.
Animal Health Component
30%
Research Effort Categories
Basic
60%
Applied
30%
Developmental
10%
Goals / Objectives
We hypothesize that certain broad-spectrum fungicides disrupt the resident microbial community on plants and decrease the ability of the community to suppress pathogen activity once the pathogen-suppressing activity of the fungicide has dissipated. We will test this hypothesis through pursuit of the following goals:Identify fungicides that result in disease resurgence in turfgrass.Identify key rhizosphere and foliar microbial community parameters, metabolites, and processes that are impacted by fungicides that result in disease resurgence.Determine whether the turfgrass-dollar spot pathosystem can serve as an appropriate model to investigate disease resurgence in a major cereal crop.
Project Methods
Objective 1: In Years 1 through 3 of the project, we will conduct a field study to determine what other common fungicides result in dollar spot resurgence in turfgrass. The study will be conducted at both the OJ Noer Turfgrass Research Facility in Madison, WI and Pleasant View Golf Course in Middleton, WI on a 'Penncross' creeping bentgrass stand maintained under putting green and fairway conditions, respectively. The treatments will consist of a non-treated control and 13 fungicide chemistries with a range of chemical structures, years on the market, and spectrum of activity (Table 1). All 14 treatments will be arranged in a randomized complete block with four replications and an individual plot size (experimental unit) of 0.9 x 3.0 m. Treatments will be randomized within each replication (block) to account for spatial differences in soil type or other micro-environmental factors. Surrounding each treatment will be a 0.5 m gap of non-treated turfgrass to ensure that no fungicide contamination occurs between treatments. Objective 1 is a field assessment of dollar spot severity in response to applications of fungicides. Dollar spot severity will be assessed by counting infection foci every two weeks between June 1st and November 1st in each year. There will be a particular focus on September and October rating dates after the cessation of fungicide application to assess the impact of fungicide on disease resurgence. We will use a three-way mixed model analysis of variance (ANOVA) with site, fungicide treatment, rating date, and their interactions as factors to determine which fungicide chemistries and which mixtures result in disease resurgence following their final application. When appropriate, treatment means will be compared with Tukey's Honestly Significant Difference (HSD) test (Tukey 1949).Objective 2: The results collected in Objective 1 in Year 1 will determine which fungicide treatments will be sampled for Objective 2 in Years 2 through 4. In Year 2 and 3, samples will be taken immediately before the initiation of fungicide applications in mid-May, in mid-July following at least two applications of each fungicide treatment, and in mid-October two months after the final fungicide application. In Year 4 the three samplings will occur on the same dates as the prior two years but no fungicides will be applied so the samplings will be done to evaluate the residual treatment effect and the recovery more than one year following the final fungicide application. Soil and foliar tissue will be randomly sampled from each plot using a 2.5-cm diameter soil probe and scissors, respectively. Ten cores will be sampled from random locations within each plot to a depth 10 cm. Adhering soil will be manually collected, operationally defined as rhizosphere soil, and the soil from the ten cores within each plot will be homogenized. Plant tissue will be collected by clipping all green plant tissue above the thatch using a sterile scissors.Rhizosphere and foliar microbial community composition will be assessed using methods that are that are consistent with the Earth Microbiome Project standard protocols. Briefly, total DNA will be extracted from rhizosphere soil and foliar tissue using the DNeasy Powersoil Pro DNA Isolation Kit (Qiagen, Germantown, MD, USA). We will target the 16S rRNA gene with widely used PCR primers (515FB, 806RB) and a Peptide Nucleic Acid (PNA) clamp to prevent chloroplast contamination in the foliar samples (Viquez at al., 2020). The fungal internal transcribed spacer (ITS2) region will be amplified using primers fITS7 and ITS4 (White et al. 1990; Ihrmark et al. 2012). Amplicons (including appropriate Illumina barcoded adapters) will be quantified using the PicoGreen dsDNA assay (ThermoFisher Scientific, Waltham, MA) and pooled together in equimolar concentrations for sequencing on an Illumina MiSeq machine at the UW-Madison Biotechnology center using 2 × 250 bp and 2 × 300 bp chemistry for 16S and ITS amplicons, respectively. Paired end sequences will be merged, de-multiplexed, stringently quality filtered (Bokulich et al. 2013), and amplicon sequence variants (ASVs) will be picked using the DADA2 pipeline in QIIME2 (Boylen et al. 2019). Taxonomy will be assigned to ASVs using the RDP classifier (Cole et al. 2014) against the most recent version of the SILVA (Quast et al. 2013) or UNITE (Nilsson et al. 2019) databases for 16S rRNA and ITS2 gene sequences, respectively.In objective 2, to test hypotheses related to univariate parameters like C mineralization rate, microbial diversity or metabolite concentration, the effects of location, fungicide, time, and their combined interaction will be determined by three-way mixed model ANOVA and means will be compared with a Tukey's HSD test. To test hypotheses related to multivariate parameters like microbial community composition and metabolomic fingerprint will be explored using Principal Component Analysis (PCA) and permutational multivariate analysis of variance (PerMANOVA; Anderson 2001). Additionally, Mantel tests (Mantel and Valand 1970) will be used to test for associations between microbiome and metabolome parameters and soil variables. Hierarchical clustering analysis of microbiome and metabolomic data will be used to determine the similarity in microbiome and metabolomic signatures across plants, soils, locations and treatments. To test for associations between microbial taxa, metabolite concentrations and disease resurgence, network analysis will be employed. We have a statistician on our team (Co-PI Solis-Lemus) to ensure the quality of statistical tests and subsequent interpretation.Objective 3: We will replicate the field trial proposed in Objective 1 in a no-till corn cropping system. Co-PI Jones will apply a similar list of treatments from Table 1 to field corn at the Arlington Agricultural Research Station in Arlington, WI. A corn hybrid frequently planted in south central Wisconsin will be selected to represent field conditions common to the area. The only difference in the treatment list for Objective 3.1 on corn will be the removal of any active ingredients, such as iprodione, that are not labeled for use on corn. Corn will be grown in a no-tillage field with a previous crop of corn silage or soybean, and a planted population of 36,500 plants per acre equivalent. The experimental design will be a randomized complete block design with four replications. Individual plots will be four rows wide and 12 meters long with a 1.5 m non-treated alley between treatments. All fungicides will be applied twice at the V6 (6 fully-emerged leaves) and VT (Tassel emergence) growth stages at label rates to all 4 rows of plot. The experimental unit will be the center two rows while the outside rows will act as the sites for destructive sampling in Objective 3.2. All disease and yield data will be recorded on the central 2 rows. Foliar diseases on corn vary from year to year so all foliar diseases present will be assessed, which in Wisconsin may include gray leaf spot (caused by Cercospora zeae-maydis), northern corn leaf blight (caused by Setosphaeria turcica), and/or tar spot (caused by Phyllachora maydis) depending on the environmental conditions that growing season. We will compare disease development and yield on treated plots to the non-treated control to determine any disease resurgence effect in corn. The study will be conducted over one growing season in year 2 and replicated in year 3. Objective 3 repeats many of the same experiments on corn, so the same statistical analyses described on turfgrass will be repeated for objective 3 on corn. Corn yield data will be corrected for harvest moisture conditions, and when applicable, yield adjustments made for final harvested plant population. When applicable, P-values will be corrected for multiple comparisons using the Benjamini and Hochberg False Discovery Rate correction (Benjamini and Hochberg 1995).