Fungicide use intensity influences the soil microbiome and fungal disease suppressiveness in amenity turfgrass

doi:10.21203/rs.3.rs-4725984/v1

Background

Understanding the factors that facilitate disease suppressive soils will contribute to more sustainable plant protection practices. Disease suppressive soils have been documented in many economically important crops, but not in turfgrass, one of the most intensively managed plant systems in the United States. Dollar spot, caused by the fungus Clarireedia jacksonii, is the most economically important disease of managed turfgrass and has historically been controlled through intensive use of fungicides. However, previous anecdotal observations of lower dollar spot severity on golf courses with less intensive fungicide histories suggests that intensive fungicide usage may suppress microbial antagonism of pathogen activity. This study explored the suppressive activity of transplanted microbiomes against dollar spot from seven locations in the Midwestern U.S. and seven locations in the Northeastern U.S. with varying fungicide use histories. Creeping bentgrass was established in pots containing homogenized sterile potting mix and field soil and inoculated with C. jacksonii upon maturity. Bacterial and fungal communities of root-associated soil and phyllosphere were profiled with short-amplicon sequencing to investigate the microbial community associated with disease suppression.

Results

The results clearly showed that plants grown in the transplanted soil microbiome collected from sites with lower fungicide intensities exhibited reduced disease severity. Plant growth promoting and pathogen antagonistic microbes may be responsible for disease suppression, but further validation is required. Additional least squares regression analysis of the fungicides used at each location suggested that contact fungicides such as chlorothalonil and fluazinam had greater influence on the microbiome disease suppressiveness than penetrant fungicides. Potential organisms antagonistic to Clarireedia were identified in the subsequent amplicon sequencing analysis but further characterization and validation is required.

Conclusion

Given the current reliance on fungicides for plant disease control, this research provides new insights into potential non-target effects of repeated fungicide usage on disease suppressive soils. It also indicates that intensive fungicide usage can decrease the activity of beneficial soil microbes. The results from this study can be used to identify more sustainable disease management strategies for a variety of economically important and intensively managed pathosystems.

Disease suppressive soils have been of great interest for decades for their ability to suppress plant diseases without the intensive use of chemical inputs [1, 2]. Suppressive soils have been identified for numerous economically important crops such as potato, wheat, strawberry, and banana [3–6]. According to the specificity and mechanism of the disease suppression, disease suppressive soils are commonly classified as either general or specific [7]. Specific suppression gains its suppressiveness from population level antagonistic microbes against one or a small number of plant pathogens, while general suppression is derived from a complex interaction of diverse microbial taxa that is often suppressive to a broader range of plant pathogens [8]. The formation of specific suppressive soils is commonly attributed to the coevolution of selective beneficial microbes, particularly those that inhibit pathogenic growth or development [7]. For example, enrichment of fluorescent Pseudomonas spp. that produce the antifungal metabolite 2,4-diacetylphloroglucinol in the wheat rhizosphere after long-term monoculture led to suppression of take-all of wheat (Gaeumannomyces graminis var. tritici) [9, 10].

Disease suppressive soils can be induced and modulated through management practices, such as long-term monoculture, crop rotation, fertilization, inoculation of pathogen antagonistic microbes, and soil amendment application [9, 11–13]. The key to management impact on the disease suppressiveness lies in manipulation of the microbial association and dynamics surrounding the host plant(s) [8, 14]. Fungicide applications are among the most common disease control methods but their impact on disease suppressive soils has not been examined for the effects on disease suppressive soil induction. Given the impact that fungicides have on the plant and soil microbiome, it is likely that fungicide usage plays a critical role in disease suppressive soil formation [15–19]. The turfgrass agroecological system is one of the most widespread and intensively managed plant systems and serves as an excellent model to study the roles of pesticide usage on microbe-plant-pathogen interactions and the resulting disease suppressiveness as they are perennial and constantly challenged by diseases.

Unlike many other economically important crops, disease suppressive soils have not been documented in turfgrass, one of the most intensively managed plant systems in the U.S. Dollar spot is caused by the fungal pathogen Clarireedia spp. and is the most economically important disease of amenity turfgrass in temperate climates [20, 21]. The fungus causes roughly circular patches of tan or brown turf 2 to 5 cm in diameter that results in a largely unplayable recreational surface [22]. Host resistance to dollar spot among cultivars of creeping bentgrass (Agrostis stolonifera) exists but has not been widely implemented as a control strategy, and cultural practices do not typically provide commercially acceptable levels of dollar spot control [23, 24]. This has resulted in more fungicides being used to suppress dollar spot than any other disease of golf course turfgrass [20]. This heavy reliance on fungicides for acceptable control has resulted in the widespread development of fungicide resistance [25], economic hardship for many golf facilities [26], and concern over human and environmental contamination [27, 28]. These concerns make the development of more sustainable dollar spot management strategies an important aspect of improving the overall sustainability of golf course management.

The authors are unaware of any empirical record of disease suppressive soils identified in turfgrass or golf course management. However, anecdotal observations by the authors of decreased dollar spot severity on multiple golf courses following the conversion to reduced-fungicide disease management programs. There has been strong interest in implementing biological control of Clarireedia using commercially available antagonistic microbes, however success in the field has been limited [29–33]. Modulating the in-situ microbiome to induce disease suppressive soils requires comprehensive foundational knowledge on the amenity turfgrass microbiome and the impacts that various management practices can have. Several studies have examined the impact of management practices on the turfgrass microbiome. Amending turfgrass soil with different forms of carbon (C) and nitrogen (N) inputs stimulated short-term pulse of enzyme activities and microbial community spikes related to C and N cycling and respiration [34]. However, N Stacey, R Lewis, J Davenport and T Sullivan [35] found no shift in the turfgrass soil microbiome in a two-year compost amendment experiment. JR Doherty and JA Roberts [36] reported that the fungicides propamocarb, fosetyl-Al, and cyazofamid did not significantly impact the rhizosphere bacterial diversity of creeping bentgrass in a two-year field study. Rhizosphere bacterial communities associated with turf grown in higher soil iron content were found to significantly alter dollar spot development and severity [37]. This last finding suggests that turfgrass microbial communities play an important role in the development of dollar spot, which may indicate that disease suppressive soils are a plausible method for suppressing dollar spot.

This study sampled 8 golf courses from the Midwestern and Northeastern U.S. with self-reported levels of high or low pesticide-use intensity. Soil from an agricultural, prairie, and forest site were also sampled in both geographic regions. Next, the connection between dollar spot suppression, past fungicide use intensity, and the microbial factors associated with disease suppression were investigated. To accomplish this, turfgrass was established in a controlled environment in potting media with soil collected from each of the sites described above. The rhizosphere and phyllosphere microbiomes were profiled using high-throughput amplicon sequencing both prior to and after inoculation with the dollar spot fungus. We hypothesized that past fungicide use intensity would impact the natural disease suppressiveness of the soil, with important implications for biological management of dollar spot and numerous other diseases in both turfgrass and other cropping systems.

Sample collection, experimental design and potting preparation

Field soil was collected between Aug 27 and Nov 1, 2019, from 14 sites in the Midwest (MW) and Northeast (NE) of the U.S. Soil was sampled from four golf courses, one agricultural field, one prairie, and one forest floor soil in each geographical region. Two golf courses in each region were identified as low fungicide intensity and two were identified as high fungicide intensity based on prior knowledge of their disease management programs (Table 1). Soils were sampled using a 2.75-cm diameter soil probe to a depth of 10 cm. Five soil cores were sampled from each site. For each golf course site, soil cores were sampled from creeping bentgrass fairways. The soil cores from MW were stored in -80°C within 3 hours of sampling and the soil cores from the NE were sampled by the field managers on-site, individually wrapped in aluminum foil, and shipped overnight to Madison, WI and immediately stored at -80˚°C upon receipt.

Table 1

Sample description and the superintendent reported nitrogen and fungicide application rate in 2019.
Name	Location	N quantity (kg/ha)	Fungicide application (a.i.-times/yr)	Fungicide quantity (g a.i./m²)	Description
Control	-	-	-	-	Sterile potting mix
MW-Ag	WI	0.00	0	0.00	Corn field
MW-Forest	WI	0.00	0	0.00	Forest floor
MW-High1	WI	35.03	14	3.13	Fairway of private golf course
MW-High2	WI	61.03	13	1.91	Fairway of private golf course
MW-Prairie	WI	0.00	0	0.00	35 plus year unmanaged prairie
MW-Low1	IL	48.82	5	0.66	Fairway of private golf course
MW-Low2	WI	24.41	9	1.75	Fairway of private golf course
NE-Ag	NY	NA	NA	NA	Apple orchard
NE-Forest	NY	0.00	0	0.00	Forest floor
NE-High1	NJ	58.59	36	6.1	Fairway of private golf course
NE-High2	NJ	82.51	22	6.43	Fairway of private golf course
NE-Prairie	NY	0.00	0	0.00	15 plus year unmanaged ecotone
NE-Low1	MA	134.27	0	0.00	Fairway of private golf course
NE-Low2	NY	89.81	6	0.85	Fairway of public golf course

‘Penncross’ creeping bentgrass seeds were surface sterilized by treating the seeds with chloroform gas overnight and exposing to UV light at 253.7 nm for two hours in a biosafety cabinet (SterilGard Model SG503A-HE, The Baker Company, Sanford, ME, USA). Surface sterilized creeping bentgrass seeds were sown and germinated on water agar before transferring to the pots to reduce the likelihood of damping off diseases injuring the seedlings. Calcined clay (Turface MVP, Oldcastle APG, Atlanta, GA, USA) and sand were homogenized (50/50 v/v) and placed in 500-ml plastic pots. The potted media was sterilized by autoclaving three times for 60 mins following placement in the pots. Soils from the field were added into the pots (5 mL/pot) with five biological replications per soil source and carefully homogenized with a sterilized spoon and aggressive shaking in close containers along with a control without field soil inoculation. A 0.5-cm layer of sterilized sand was applied on top of the transferred seedlings to keep the roots covered and avoid drying out. Each pot was wetted with sterile distilled water with a vaporizer before and after seedlings’ transfer. All the tools and containers used in contact with seed and soil were sterilized. The pots were then placed in a UVC-sterilized growth chamber (Model 136LLVL, Percival Scientific, Perry, IA, USA) at 20/18°C day/night temperature and 40% humidity with 16 hours light period for two months before pathogen inoculation. The turf was trimmed to a height of 1-cm with sterilized scissors and irrigated with sterilized ddH2O using a vaporizer at a rate of approximately 20 mL/pot every other day throughout the incubation prior to pathogen inoculation.

Pathogen inoculation and disease assessment

Inoculum was prepared by transferring one-week old PDB-cultured Clarireedia jacksonii (strain 2F92-1 collected in Madison, WI [38]) hyphae onto sterilized rye grains and incubated for 10 days at 20˚°C in the dark. Each pot was inoculated by placing five Clarireedia-infested rye grains in the middle of each pot at a depth of 0.5 cm below the turf canopy to ensure good contact between rye grains and the turf canopy-soil interface. The incubation condition was adjusted to 28/20˚°C day/night temperature and 75% humidity with 16 hours light period to encourage dollar spot development.

The disease development monitoring followed a similar procedure to Chou et al. (2021) [37]. Briefly, digital pictures were taken 30 cm vertically above the turf, and the pictures were analyzed with Fiji [39] to calculate the percentage of green pixels within the measured turf surface area (i.e. greenness). The measurements were taken every 48 hours starting from the day of pathogen inoculation and continued until 16 days after pathogen inoculation (DAI) and once again on 20 DAI. Percent greenness was defined as the greenness of each pot compared with the baseline greenness (100%) of that same pot measured on the day of pathogen inoculation (0 DAI).

Soil and phyllosphere sampling and DNA sequencing

Soil and phyllosphere samples were collected from each pot immediately prior to pathogen inoculation and again 20 DAI. For the soil samples, soil was sampled from each pot using an8-mm metal cork borer to a depth of 2.5 cm with two random soil cores taken and pooled at each sampling point. For the phyllosphere samples, turfgrass leaves were collected from each pot by trimming the turfgrass plants at 1 cm using sterilized scissors.

Microbial DNA was extracted with Qiagen DNeasy PowerSoil Pro Kits (Qiagen, Hilden, Germany) and phyllosphere DNA was extracted using a Maxwell® RSC Plant DNA Kit (Promega, Madison, WI, USA). The amplicon sequencing libraries were prepared following a modified method from MS Cox, CL Deblois and G Suen [40]. Briefly, the extracted DNA from each sample was diluted to 20 ng/uL followed by the amplification of 16S V4 and ITS2 regions using barcoded primers as described in [40]. Each 25 uL PCR reaction contained 5 uL of template DNA and 12.5 µL of NEB master mix undergone cycling conditions of an initial 95°C for 5 min followed by 25 cycles of 15 sec at 95°C, 30 sec at 60°C, 30 sec at 72°C, and then a final 72°C for 8 min before storing the amplicons in -20°C. The PCR reactions for phyllosphere samples were the same as soil samples except the water was replaced with mitochondrial and chloroplast DNA clamps at 1 µM and the cycling condition had an extra step of 68°C for 30 sec [41]. The amplicons were checked on 1.5% agarose gel, gel-extracted with Zymoclean Gel DNA Recovery Kit (Zymo Research, Irvine, CA, USA), normalized with Mag-Bind® EquiPure gDNA Normalization Kit (Omega Bio-Tek, Norcross, Georgia, USA), quantified with Qubit™ dsDNA HS assay (Thermo Fisher Scientific, Waltham, MA, USA), and equimolar pooled prior to sequencing on Illumina MiSeq (llumina, San Diego, CA, USA) system with 2×250 and 2×300 kits for 16S and ITS amplicons, respectively, at the University of Wisconsin – Madison Biotechnology Center.

Bioinformatics and data analysis

Sequencing reads were demultiplexed using the default setting of bcl2fastq (llumina, San Diego, CA, USA), quality filtered and cleaned using the DADA2 [42] pipeline to generate amplicon sequence variants (ASVs) using R 4.0.2, and the taxonomic ranks were assigned with SILVA (v138) and UNITE (v8.2) reference databases [42–44]. Only forward sequences were used for ITS sequences due to low quality reverse reads. Principal coordinates analysis (PCoA), ASV richness, Shannon diversity and permutate-multivariate analysis of variance (PERMANOVA) were performed with package “vegan” and “phyloseq” in R [45, 46]. Due to the large number of variables when using ASVs as predictors for dollar spot severity, a Random Forest machine learning algorithm was used to build the prediction model instead of conventional linear regression. The model optimization procedure followed the description outlined in PBd Costa, GMN Benucci, M-Y Chou, JV Wyk, M Chretien, G Bonito and BG Turgeon [47]. Briefly, Boruta feature selection was performed using the “Boruta” package with 100 iterations, 999 permutations, and 100 loops to identify the potential important variables and then the relative abundances of the consensus ASV were used in Random Forest (RF) modeling with the “randomForest” package in R [48, 49]. Fungicide effect on the turfgrass greenness was modeled using Partial Least Squares Regression with “pls” package in R [50] as the independent variables were all significantly correlated with each other and had a non-linear relationship with the greenness. The model p-value was derived by calculating the chance of the max r square values of the permutate models equal or greater than the minimum r-square value of the current model with 10,000 permutations.

Disease development

Significant differences in dollar spot development, as measured by turf greenness, were observed (Fig. 1). Disease symptoms first appeared 6 DAI and significant differences between treatments began to appear between 8 and 10 DAI. Although minor disease progression was observed post 16 DAI, the disease symptoms were most severe at the end of the experiment at 20 DAI (Fig. 1). By the end of the incubation, the non-treated control, which had no field soil added to the potting media, had the most severely diseased turf with less than 10% turf greenness. Other treatments exhibiting significant disease included NE-Prairie (18.8%), MW-Prairie (28.23%), and NE-High2 (34.64%). The treatments exhibiting the least amount of disease were NE-Low1 (78.14%) followed by MW-Low2 (77.87%), and NE-Forest (77.13%).

Microbial richness and diversity

A total of 6,344,855 and 13,340,254 reads were yielded with an average of approximately 16,000 and 33,000 reads after initial quality filtering for 16S and ITS samples, respectively. Bacterial and fungal community composition was distinct among sample types (soil and phyllosphere) and sampling stage (pretransplant, immediately before Clarireedia inoculation, and at peak of disease) as visualized in the two-dimension Principal Coordinate Analysis (PCoA) with Bray-Curtis distance (Fig. 2a, 2b). When analyzing the soil and phyllosphere samples separately, sample clustering by treatment was clearly observed for both bacterial and fungal communities regardless of sampling stage (Fig. 2c,d,e,f). Notably, shifts in soil bacterial and fungal communities occurred for all treatments after field soil microbiome transplantation. Also, there seemed to have a clear phyllosphere bacterial community difference among sampling stages (Fig. 2E). The visual observation was statistically confirmed by permutational multivariate analysis of variance (PERMANOVA) (Table 2) and paired-PERMANOVA (Table S2). For both bacterial and fungal communities, the sample type and stage, treatments, and the interactions all significantly explained the microbiome variances (Table 2). For the soil community, sampling type and stage explained the most variance for bacterial communities (R² = 0.25, p < 0.0001) followed by treatment (R² = 0.153, p < 0.0001), whereas treatment effect was more prevalent in fungal communities (R² = 0.197, p < 0.0001) than that of sampling type and stage (R² = 0.145, p < 0.0001). For the phyllosphere, treatment always explained the most variance for bacterial (R² = 0.299, p < 0.0001) and fungal (R² = 0.318, p < 0.0001) communities. Although the read number variation across the samples was also a significant factor, the variance explained ranging from 2.1 to 11% were only a fraction of the other effects. The treatment effect was further validated with paired-PERMANOVA where almost all pairs across sample types and sampling time were significantly different in both bacterial and fungal communities with rare exceptions (Table S1). ASVs homogeneity test was conducted with beta-dispersion, and only significant differences were observed in fungal communities among different field soils whereas other sample types and sampling stages were not significant across treatments (Table S2).

Table 2

PERMANOVA analyses for turf phyllosphere and rhizosphere soil bacterial and fungal communities. Reads refer to number of reads after quality filtering for each sample, treatments represent different sources of field soil inocula, and TypeStage indicates the sample types including phyllosphere and rhizosphere soil as well as sampling stages including field inocula, pre-inoculation of pathogen, and peak of disease.
Overall 16S	R²	Pr(> F)		Overall ITS	R²	Pr(> F)
Reads	0.04838	1.00E-04	***	Reads	0.03713	1.00E-04	***
TypeStage	0.24995	1.00E-04	***	TypeStage	0.14481	1.00E-04	***
Treatment	0.15293	1.00E-04	***	Treatment	0.1968	1.00E-04	***
Reads:TypeStage	0.03253	1.00E-04	***	Reads:TypeStage	0.03215	1.00E-04	***
Reads:Treatment	0.04885	1.00E-04	***	Reads:Treatment	0.05584	1.00E-04	***
TypeStage:Treatment	0.22674	1.00E-04	***	TypeStage:Treatment	0.21946	1.00E-04	***
Reads:TypeStage:Treatment	0.06748	1.00E-04	***	Reads:TypeStage:Treatment	0.07556	1.00E-04	***
Residuals	0.17314			Residuals	0.23824
Soil 16S	R²	Pr(> F)		Soil ITS	R²	Pr(> F)
Reads	0.04743	1.00E-04	***	Reads	0.02177	0.0001	***
Stage	0.14668	1.00E-04	***	Stage	0.11756	0.0001	***
Treatment	0.27599	1.00E-04	***	Treatment	0.23354	0.0001	***
Reads:Stage	0.01523	1.00E-04	***	Reads:Stage	0.01483	0.0001	***
Reads:Treatment	0.06956	1.00E-04	***	Reads:Treatment	0.07179	0.0001	***
Stage:Treatment	0.17918	1.00E-04	***	Stage:Treatment	0.20455	0.0001	***
Reads:Stage:Treatment	0.05529	1.00E-04	***	Reads:Stage:Treatment	0.06396	0.0049	**
Residuals	0.21064			Residuals	0.272
Phyllosphere 16S	R²	Pr(> F)		Phyllosphere ITS	R²	Pr(> F)
Reads	0.08154	1.00E-04	***	Reads	0.11228	1.00E-04	***
Stage	0.15405	1.00E-04	***	Stage	0.08958	1.00E-04	***
Treatment	0.29923	1.00E-04	***	Treatment	0.31829	1.00E-04	***
Reads:Stage	0.02079	1.00E-04	***	Reads:Stage	0.0139	1.00E-04	***
Reads:Treatment	0.09847	1.00E-04	***	Reads:Treatment	0.12777	1.00E-04	***
Stage:Treatment	0.11546	1.00E-04	***	Stage:Treatment	0.07731	1.00E-04	***
Reads:Stage:Treatment	0.0543	1.00E-04	***	Reads:Stage:Treatment	0.05182	1.00E-04	***
Residuals	0.17616			Residuals	0.20905

Differences in microbiome α-diversity, measured as natural log richness and Shannon diversity index, between field soil inoculum were observed (Fig. 3). For bacterial richness, NE-Low2, NE-Forest, and MW-Prairie were among the lowest, and NE-High2 and MW-Low2 were among the highest (Fig. 3a). This trend generally held for bacteria in the soil at pathogen inoculation and became more even at peak disease. In contrast, the phyllosphere bacterial richness was even across the treatments except NE-Prairie and MW-Ag, but became more divergent at the peak of disease. For fungal richness in the field soil inoculum, non-golf course soil from Midwest, NE-High1, NE-High2, NE-Low1, and NE Prairie had the highest richness and the MW-High1 was the lowest (Fig. 3b). The soil fungal richness decreased after soil inoculation and turf establishment, but MW-Ag and MW-Forest were among the highest in richness and MW-High1 was among the lowest, which was similar to the bacterial samples. The soil fungal richness became slightly more divided among the treatments at the peak of disease compared to pathogen inoculation sampling, while the phyllosphere fungal richness was more even across the treatments at the end of the experiment when the disease peaked. The potting media without soil had the lowest bacterial and fungal richness in the soil regardless of the sampling stage. Shannon diversity generally showed a similar trend as the richness (Fig. S1).

Identifying potential disease suppressive predictor microbes

Microbial taxa in the turf-associated microbiome showed different relative abundances across the treatments (Fig. S2 and S3). However, due to the complex microbial composition and a large number of variables to model the microbial-disease suppressive relationship, a machine learning algorithm was used to decipher the association. Boruta was applied to select the relevant bacterial and fungal ASVs, and Random Forest was used to build a predictive model. Significant models were built with more than 60% variance explained by using either soil bacterial or fungal ASVs to predict disease suppressiveness, whereas models built with phyllosphere bacterial and fungal ASVs resulted in 25.15% and 53.82% variance explained, respectively (Fig. S4). All models were significant (model P-value < 0.05), suggesting non-random microbial assembly that was influenced by treatment. Top disease suppressiveness bacterial and fungal predictors were selected and ranked by their increase in mean square error for each sample type and sampling time (Fig. 4). Distinctively effective predictor ASVs, as evaluated by increase of mean square error, were observed for each sample type and stage including the bacteria Gaiella sp., Methylocella sp., Stenotrophomonas rhizophila, Neorhizobium galegae, Pantoea ananatis, and the fungi Arthrinium malaysianum and Cladosporium sphaerospermum.

For the RF model-selected important microbial predictors, correlation analyses were performed to associate their relative abundance and the dollar spot disease suppressiveness (Table 3). The RF model selected ASVs from samples at the time of pathogen inoculation and the results showed many significantly and positively correlated taxa, including the fungi Microdochium neoqueenslandicum, Mucor moelleri, Saitozyma podzolica, Microdochium sp., Penicillium simplicissimum, Chaetomium homopilatum, Solicoccozyma terricola (Table 3a), and the bacteria Mesorhizobium ciceri, Bradyrhizobium elkanii, unidentified Xanthobacteraceae, and Phenylobacterium sp. (Table 3b).

Table 3. Correlation test of relative abundances of random forest selected important (a) bacterial and (b) fungal taxa from the turf rhizosphere soil prior to Clarireedia inoculation with turfgrass greenness after disease development, and field nitrogen and fungicide applications.

(a)

	Greenness (20 DAI)		N quantity		Fungicide quantity		Fungicide frequency
Fungal taxa	R	P-value	R	P-value	R	P-value	R	P-value
Microdochium neoqueenslandicum	0.18	0.01	0.23	0.00	-0.14	0.05	-0.15	0.05
Penicillium ochrochloron	-0.11	0.13	-0.15	0.05	-0.12	0.09	-0.13	0.06
Cladosporium sphaerospermum	-0.15	0.05	-0.10	0.23	-0.12	0.08	-0.11	0.10
Gibberella zeae	-0.35	0.00	-0.42	0.00	-0.31	0.00	-0.30	0.00
Mucor moelleri	0.24	0.00	0.08	0.31	-0.24	0.00	-0.23	0.00
Saitozyma podzolica	0.27	0.00	-0.20	0.01	-0.43	0.00	-0.43	0.00
Microdochium spp.	0.14	0.05	0.04	0.61	0.16	0.03	0.17	0.02
Unidentified Hypocreales	-0.26	0.00	-0.19	0.01	-0.21	0.00	-0.20	0.01
Fusarium spp.	-0.25	0.00	-0.08	0.31	-0.16	0.03	-0.17	0.02
Penicillium brasilianum	-0.11	0.14	-0.31	0.00	-0.25	0.00	-0.26	0.00
Sarocladium kiliense	-0.33	0.00	-0.19	0.01	-0.15	0.04	-0.16	0.03
Penicillium simplicissimum	0.20	0.01	0.09	0.30	0.08	0.21	0.07	0.31
Chaetomium homopilatum	0.44	0.00	0.03	0.67	-0.33	0.00	-0.32	0.00
Penicillium simplicissimum	-0.02	0.77	-0.14	0.07	-0.13	0.07	-0.13	0.07
Staphylotrichum spp.	-0.04	0.64	0.01	0.83	0.21	0.00	0.24	0.00
Arthrinium malaysianum	-0.14	0.05	-0.25	0.00	-0.13	0.07	-0.13	0.07
Staphylotrichum spp.	-0.05	0.50	0.06	0.46	0.24	0.00	0.27	0.00
Solicoccozyma terricola	0.28	0.00	-0.24	0.00	-0.46	0.00	-0.46	0.00
Stachybotrys chartarum	-0.18	0.01	-0.14	0.06	-0.11	0.11	-0.11	0.10

(b)

	Greenness (20 DAI)		N quantity		Fungicide quantity		Fungicide frequency
Bacterial taxa	R	P-value	R	P-value	R	P-value	R	P-value
Paenibacillus spp.	0.00	0.98	0.11	0.17	-0.02	0.84	0.00	0.95
Novosphingobium resinovorum	-0.35	0.00	-0.18	0.01	-0.15	0.07	-0.13	0.10
Massilia spp.	-0.25	0.00	-0.09	0.24	0.16	0.06	0.15	0.07
Arthrobacter alpinus	-0.18	0.02	0.03	0.64	0.25	0.00	0.23	0.00
Mesorhizobium ciceri	0.15	0.04	0.11	0.17	0.14	0.08	0.12	0.12
Sphingobium spp.	0.10	0.15	0.05	0.57	0.14	0.08	0.14	0.10
Paenibacillus agarexedens	-0.26	0.00	-0.34	0.00	-0.16	0.06	-0.16	0.07
Methylocella spp.	-0.23	0.00	-0.43	0.00	-0.24	0.00	-0.26	0.00
Unidentified Fibrobacteraceae	-0.16	0.03	-0.11	0.17	0.03	0.68	0.05	0.50
Bradyrhizobium elkanii	0.29	0.00	0.23	0.00	0.21	0.01	0.19	0.02
Ancylobacter spp.	-0.34	0.00	-0.40	0.00	-0.36	0.00	-0.36	0.00
Novosphingobium resinovorum	-0.06	0.41	0.09	0.23	0.08	0.31	0.09	0.29
Unidentified Xanthobacteraceae	0.24	0.00	0.24	0.00	-0.01	0.84	-0.01	0.94
Paenibacillus alginolyticus	-0.01	0.91	-0.02	0.82	0.08	0.34	0.06	0.44
Fontimonas spp.	-0.16	0.03	-0.21	0.01	-0.07	0.34	-0.08	0.34
Bdellovibrio spp.	-0.29	0.00	-0.17	0.03	-0.07	0.37	-0.07	0.40
Georgfuchsia spp.	-0.22	0.00	-0.15	0.05	-0.12	0.11	-0.12	0.14
Sphingopyxis macrogoltabida	-0.14	0.05	0.05	0.54	0.13	0.09	0.13	0.10
Phenylobacterium spp.	0.14	0.05	0.24	0.00	0.14	0.08	0.15	0.07

Correlation analysis showed a significant negative correlation between fungicide application intensity and dollar spot suppressiveness where both fungicide application quantity (g/m²) (R= -0.72, p < 2.2e − 16) and frequency (sum of a.i. × times) (R= -0.71, p < 2.2e − 16) were significant (Fig. 5). Total N application did not significantly correlate with dollar spot suppressiveness (R= -0.011, p = 0.9) (Fig. S5). Specific fungicides and fungicide classes were examined for their relationship with dollar spot microbial suppressiveness using partial least squares regression (PLSR). The PLSR model suggested effective prediction of turfgrass greenness at 20 DAI using all five fungicides or fungicide classes commonly used at the sampling sites (p-value = 8e-4). Fluazinam, among all fungicides, had the most predictive power followed by chlorothalonil, demethylation inhibitor (DMI) fungicides, dicarboximide fungicide, and succinate dehydrogenase inhibitor (SDHI) fungicides (Fig. S6).

Many important disease suppressive microbial predictors were significantly correlated with the fungicide and nitrogen application intensity (Table 3). The significantly correlated microbes were generally shared between fungicide application frequency and quantity. For fungicide application intensity, the fungi Gibberella zeae, Mucor moelleri, unidentified Hypocreales, Fusarium spp., Penicillium brasilianum, Sarocladium kiliense, Chaetomium homopilatum, Solicoccozyma terricola (Table 3a), and the bacteria Methylocella spp., Ancylobacter spp. were negatively correlated (Table 3b). In contrast, the fungi Microdochium spp., Staphylotrichum sp., and the bacteria Arthrobacter alpinus, Bradyrhizobium elkanii were positively correlated with fungicide application intensity (Table 3a). Although nitrogen application did not correlate with disease suppressiveness, significant correlation with relative abundances of individual taxa were observed. Fungi Microdochium neoqueenslandicum, bacteria Bradyrhizobium elkanii, unidentified Xanthobacteraceae, and Phenylobacterium spp. positively correlated with the N application, whereas fungi Gibberella zeae, Saitozyma podzolica, unidentified Hypocreales, Penicillium brasilianum, Sarocladium kiliense, Arthrinium malaysianum, Solicoccozyma terricola and bacteria Novosphingobium resinovorum, Paenibacillus agarexedens, Methylocella spp., Ancylobacter spp., Fontimonas spp., Bdellovibrio spp., Georgfuchsia spp. were negatively correlated with the N application (Table 3).

The presence of disease suppressive soil in turfgrass

We observed differences in microbiome-mediated dollar spot suppressiveness among soils collected from 14 locations across the Midwest and Northeast U.S. encompassing golf courses, agricultural, and native prairie landscapes. To our knowledge this is the first description of disease suppressive soils in turfgrass. We also demonstrated that the suppressive ability of the soil could be transplanted to a sterile potting media, which is typically a key aspect of specific suppressive soils and may provide future directions for research in turfgrass and other agricultural and horticultural pathosystems. Conferring disease suppression by transplanting a disease suppressive soil into a conducive soil has been observed in several plant pathosystems including Rhizoctonia solani in sugar beet [51], and Ralstonia solanacearum in eggplant and tomato [52, 53]. In all these studies the primary mechanism for disease suppression was found to be the retention and enrichment of microbes antagonistic to the pathogen. Although this is yet to be verified in our study, the observed disease suppression translatability may suggest the enrichment of plant growth promoting microbes and microbes antagonistic to Clarireedia.

Another interesting finding is that turf grown with the forest soil microbiome had a lower dollar spot severity than turf grown with the prairie soil microbiome. In fact, turf transplanted with the prairie microbiome had higher dollar spot severity than all other treatments except for one (NE-High2), suggesting potential dysbiosis after transplant of prairie soil microbiome. It is also possible that the Clarireedia had better fitness in the prairie-associated microbial community than that of forest as sampled prairies were likely to have more colonization of Poaceae species, which is phylogenetically close to the hosts of Clarireedia. Additionally, forest soil may harbor microbes that suppress fungal pathogens or promote creeping bentgrass defense against dollar spot. Multiple studies have found forest soil harboring antifungal compound producing microbes or those that can induce systemic disease resistance [54–56].

Fungicide intensity impacts dollar spot suppressiveness

There was a clear inverse relationship in our study between fungicide intensity at the sampling location and dollar spot suppressiveness in the inoculated pot assay. Furthermore, the use of contact fungicides including fluazinam and chlorothalonil seemed to have more impact on the microbial Clarireedia suppressiveness as they were identified as top turf greenness predictors at 20 DAI. Specific and induced disease suppressive soil formation depends largely on the soil management and favors management that allows for the growth, selection, and coevolution of the plant, antagonistic microbes, and the pathogen in a monocropping system over a relatively long time scale [7]. Past research has found that amendments that enriched the soil microbial diversity, such as organic soil amendments, were found to effectively confer disease suppression [57, 58]. In turfgrass, compost applications were shown to suppress dollar spot in turfgrass with a postulated mechanism of reduced pathogen fitness due to microbial competition [59]. However, the role of fungicide applications in suppressive soil formation has remained unclear. This present study demonstrates that reduced fungicide intensity can also induce disease suppressive soil as it potentially allows a more diverse soil microbiome to grow and facilitate the plant-microbe and microbe-microbe coevolution between plant, antagonistic microbes, and the pathogen. More specifically, fewer chemical inputs contributing to increased dollar spot suppressiveness supports the proposed theoretical framework of induced suppressive soil formation in response to pathogen activity [8].

Golf courses with lower fungicide intensity generally had higher levels of dollar spot suppression. However, some golf courses with lower fungicide intensity failed to sustain dollar spot suppression after microbiome transplanting throughout the entire controlled environment study. This likely is an indicator of natural, rather than induced, disease suppression which is largely dependent on the soil physical and chemical properties [7]. In addition, loss of microbial diversity, especially for the fungal community, due to freezing of field soil inocula in storage may have also led to failed disease suppression in some of the treatments and may explain why fungal predictors accounted for less variance in the RF suppressive soil predictive model. Another possibility may be that the controlled environment and the newly established turf in our study provided low fitness for the key plant beneficial and pathogen antagonistic microbes to colonize and thrive, thus failing to provide the plant beneficial functions. Nevertheless, with the observed link between field fungicide usage and induced disease suppression in a controlled environment, this study reveals the essential role of fungicide application in disease suppressive soil formation.

Potential dollar spot inhibitory and turf health promoting microbes in the soil

The RF models identified several microbes that were positively correlated with dollar suppressiveness and negatively correlated with fungicide intensity at pathogen inoculation. These included fungi such as Mucor moelleri, Chaetomium homopilatum, and Solicoccozyma terricola. Mucor moelleri has been shown to promote plant growth through antagonistic activity against the fungal pathogens Athelia rolfsii and Colletotrichum gloeosporiodes in both infested tomato plants and using in vitro assays (Nartey et al 2021). Many species in the genus Chaetomium were previously found to produce diverse bioactive compounds including many antibiotics [60] and have been suggested as a potential biocontrol agent against a broad spectrum of plant oomycetes and fungal pathogens such as Phytophthora spp. in durain, black pepper, tangerine, Fusarium oxysporum in tomato, and Sclerotium rolfsii in corn [61]. Solicoccozyma terricola is linked to soil biomass degradation and was previously found to enrich in Streptomyces lydicus M01-treated soil for Alternaria leaf spot suppression in cucumber [62].

Bacteria identified by the RF models that were positively correlated with dollar spot suppression were Mesorhizobium ciceri, Bradyrhizobium elkanii, unidentified Xanthobacteraceae, and Phenylobacterium spp. Among the four bacteria that were positively correlated with higher dollar spot suppressiveness, two of them were root-nodulating bacteria that have proven plant growth promotional effects. Mesorhizobium ciceri was found to facilitate nutrient acquisition and also alleviate the negative effect of fungicide kitazin on Chickpea (Cicer aritienum L.) with reduced oxidative damage and cell death (Shahid 2021). Bradyrhizobium elkanii is a well-studied symbiont with many legume species that fixes atmospheric N into plant available N and facilitates increased plant growth [63, 64], and was found to increase cowpea growth under water deficit scenarios [65]. The higher relative abundance of Phenylobacterium sp. was previously associated with improved barley growth, but the actual plant growth promoting mechanism remains unclear [66]. Future research on dollar spot suppressive soils and plant health promoting microbes should focus on these organisms because of their strong correlation with reduced dollar spot in this study.

Potential dollar spot inhibitory and turf health promoting microbes in the phyllosphere

The field soil transplant also contributed to the phyllosphere microbiome assembly at pathogen inoculation. Important fungal predictors of disease suppression selected by the RF model included many known antifungal compound-producing fungi such as Cladosporium sphaerospermum ASVs, Sarocladium kiliense, Microdochium spp., Penicillium simplicissimum, Staphylotrichum spp., Alternaria spp. Almost all RF-selected bacteria in the phyllosphere were previously shown to have antifungal properties, including notable ones such as Stenotrophomonas spp. and Paenarthrobacter spp.

The bacterium Stenotropphomonas rhizophila was identified as the most important phyllosphere bacterial predictor by the RF model. This bacterium was previously found to have antifungal properties, inhibited the growth of plant pathogens Alternaria alternata and Botrytis cinerea in vitro, and suppressed B. cinerea infection when sprayed on tomato leaves [67, 68]. Paenarthrobacter spp. were found to be the plausible key taxa in a Rhizoctonia solani suppressive soil in rice where in vitro experiments using the bacterial isolate from the suppressive soil confirmed the suppression activity of P. ureafaciens against fungal phytopathogens such as R. solani and Colletotrichum spp. [69]. Although there are known species of Microdochium that can cause disease in turfgrass, namely Microdochium nivale, many other Microdochium species can produce antifungal compounds such as isocoumarin derivatives and monocerin [70] and have shown biocontrol activity in barley against Gaeumannomyces graminis var. tritici [71]. Species in Staphylotrichum were found to mildly inhibit the growth of the plant pathogen Corynespora cassiicola on PDA in vitro [72] and was associated with healthy Ginseng (Panax notoginseng) in soil conducive to replant root-rot [73]. Penicillium simplicissimum can produce diverse antifungal compounds [74, 75] and induce plant systemic defense response in Cucumber (Cucumis sativus) against Colletotrichum orbiculare [76] It has also demonstrated efficacy as a biological control agent against Pythium damping-off in beetroot (Beta vulgaris) [77], Verticillium wilt in Cotton (Gossypium hirsutum) [78], and Puccinia striiformis f. sp. tritici in wheat [79]. Cladosporium sphaerospermum produces diverse polyketides that have shown biological control effects against Botrytis in vitro and on strawberry and tomato fruits in vivo [80, 81]. The red sage endophytic fungus Sarocladium kiliense has been shown to produce antifungal compounds that were confirmed with an in vitro plate assay [82].

All of these organisms were found either in higher abundances or as effective dollar spot suppression predictors in the phyllosphere of plants grown in soils from sites with reduced fungicide usage, suggesting a connection between the soil microbiome and phyllosphere that can mediate disease development. In addition, our findings also support the previous studies on soil as a source for phyllosphere microbiome assembly, which provides essential functions such as pathogen suppression [83–85].

Untargeted effect of fungicide may modulate plant beneficial microbes

The PLSR model suggested contact fungicides such as fluazinam and chlorothalonil may have more influence on the microbiome disease suppressiveness than penetrant fungicides. Fluazinam (3-chloro-N-(3-chloro-2,6-dinitro-4-(trifluoromethyl)phenyl)-5-(trifluoromethyl)-2-pyridinamine) is a pyridinamine fungicide with a suggested mode of action being inhibition of ATP synthetase in fungal cells [86]. Chlorothalonil (2,4,5,6-tetrachloroisophthalonitrile) is a chloronitrile fungicide that reacts with fungal thio-dependent enzymes and leads to cell death [87]. Both chlorothalonil and fluazinam are commonly used to control a broad array of turfgrass foliar diseases. Their impact on the turfgrass microbiome has not been investigated but was previously found to impact the microbiome of other cropping systems [18, 88–90]. In addition to the fungicidal effect of fluazinam, it was found to be highly toxic to bacteria in the luminescent bacteria toxicity test in a controlled-environment and a potato field soil that received 205 g/ha [88]. In the same study, the authors also found that the residue of fluazinam is long lasting as the transformation products can be found throughout the season and even over-winter in the soil [88]. In one study, application of fluazinam in Chinese cabbage (Brassica rapa) led to reduced fungal abundance and short-term elevated bacterial diversity and the associated catabolism functional diversity likely due to increased substrate availability from fungal death [91]. When compared to the contact fungicide mancozeb, fluazinam impacted the soil bacteria to a lesser extent but still eliminated 25 bacterial species from 0.95 L of soil received 0.054 g of Fluazinam after 6 weeks [90].

Chlorothalonil is known to have bactericidal effects and is used to control several bacterial diseases in plants suggesting its cross-kingdom broad spectrum activity [92]. Application of chlorothalonil was previously found to significantly affect soil bacterial and fungal community structure and inhibit soil dehydrogenase, catalase, and acid phosphatase activities [18, 89]. Interestingly, a recent study found that chlorothalonil suppressed the pathogen-antagonistic bacteria on amphibian (Lithobates vibicarius) skin which could potentially lead to loss of host immunity [93]. Though the results from our study suggest that fluazinam and chlorothalonil can reduce the activity of microbes antagonistic to Clarireedia spp., direct research exploring these potential effects in greater detail is warranted to understand the mechanisms behind the effects.

Other indirect effects of fungicides on soil ecosystem functions, microbial diversity nutrient cycling, and disease susceptibility have been extensively studied and reviewed in the past [66, 94]. For example, X Wang, Z Lu, H Miller, J Liu, Z Hou, S Liang, X Zhao, H Zhang and T Borch [17] showed that application of the quinone outside inhibitor (QoI) fungicide azoxystrobin inhibited the activities of urease, invertase, and phosphatase while promoting the activity of catalase in the soil. In this same study, bacterial α-diversity was reduced, and the community was restructured by azoxystrobin. Repeated fungicide applications can also reduce the colonization of arbuscular mycorrhizal fungi (AMF) [15] and decrease activity of root nodule forming bacteria [95]. In contrast, one study found that foliar application of the QoI fungicide pyraclostrobin enhanced the root nodulation and nitrogen fixation in soybean. Bacteria formed root nodules host a variety of plant beneficial microbes that release plant hormones, facilitate nutrient uptake, and many of them can also produce antifungal compounds [96], which potentially protect plants from fungal pathogens.

For the untargeted effect of fungicides specific to our study, among the bacteria indicated as important by the RF model, Ancylobacter sp. and Methylocella sp. negatively correlated with fungicide intensity. Interestingly, both bacteria negatively correlated with the fungicide intensity were methylotrophic [97, 98]. Species in the genus Ancylobacter were found to fix N and promote growth of rice [99] and can potentially regulate the ethylene signaling, providing N and phosphate to plants according to in vitro substrate utilization assays [100]. Also, many species in Ancylobacter are capable of utilizing oxalate as a carbon source [101, 102], which has been identified as one of the major virulence factors in Clarireedia [103, 104]. Therefore, Ancylobacter might be another crucial organism that played a role in suppressing dollar spot severity in this study. Collectively, these results suggest that higher fungicide application intensity may be negatively impacting plant-beneficial bacteria and fungi in the soil, leaving the turfgrass plant more susceptible to disease development.

The results presented here show a clear relationship between fungicide intensity and the dollar spot suppressive ability of a golf course soil. This has important implications not only for the development of more sustainable golf course management strategies, but also for more sustainable management of diseases in other intensively managed crops like potatoes and apples. The RF models and correlation analyses suggest that disease suppression is due to improved nutrient acquisition, protection against oxidative stress, and pathogen inhibition. Future research should focus on further work with the organisms of interest identified here to further clarify the mechanisms of their benefit and whether they would make good candidates for potential biocontrol agents. In addition, impacts of specific fungicides also need to be investigated to determine whether certain fungicides have more significant impacts on beneficial microbes than others and should be avoided. Further understanding these aspects will lead to healthier soils, improved disease control, and reduced reliance on synthetic chemicals.

Availability of data and materials

Amplicon sequence data are available in the Sequence Read Archive under BioProject accession numbers PRJNA1002783 and PRJNA1002784.

Acknowledgements

The authors thank Dr. Frank Rossi, Dr. Bruce Clarke and anonymous golf courses for coordinating the sample collection, Dr. Garret Suen and Joseph H. Skarlupka V, and University of Wisconsin-Madison Statistical Consulting Lab for their assistance in statistical analyses.

Funding

This research was funded by USGA and O.J. Noer Research Foundation.

Author information

Authors and Affiliations

Department of Plant Biology, Rutgers University, New Brunswick, NJ 08901, U.S.A.
Ming-Yi Chou

Horticulture Section, School of Integrative Plant Science, Cornell University, Ithaca, NY 14853, U.S.A.
Jenny Kao-Kniffin

Department of Plant Pathology, University of Wisconsin-Madison, Madison, WI 53706, U.S.A.
Ming-Yi Chou, Apoorva Tarihalkar Patil, Daowen Huo, Qiwei Lei & Paul Koch

Contributions

M.Y.C. conceived and performed research and data analysis, created figures and tables, and wrote the manuscript. A.T.P., D.H. and Q.L. performed research and reviewed manuscript. J.K.K. collected samples, reviewed and edited the manuscript. P.L.K. supervised the project, conceived research, wrote, reviewed and edited the manuscript.

Corresponding author

Correspondence to Paul Koch and Ming-Yi Chou

Ethics declarations

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

Competing interests
The authors declare no competing interests.

Hornby D. Suppressive Soils. Annual Review of Phytopathology. 1983;21(1):65–85; doi: 10.1146/annurev.py.21.090183.000433.
Baker K, Cook RJ. Biological control of plant pathogens. WH Freeman and Company.; 1974.
Liu D, Anderson NA, Kinkel LL. Biological control of potato scab in the field with antagonistic Streptomyces scabies. Phytopathology. 1995;85(7):827–31.
Kwak Y-S, Weller DM. Take-all of wheat and natural disease suppression: a review. The Plant Pathology Journal. 2013;29(2):125.
Kim D-R, Jeon C-W, Shin J-H, Weller DM, Thomashow L, Kwak Y-S. Function and distribution of a lantipeptide in strawberry Fusarium wilt disease–suppressive soils. Molecular Plant-Microbe Interactions. 2019;32(3):306–12.
Shen Z, Ruan Y, Xue C, Zhong S, Li R, Shen Q. Soils naturally suppressive to banana Fusarium wilt disease harbor unique bacterial communities. Plant and Soil. 2015;393:21–33.
Kinkel LL, Bakker MG, Schlatter DC. A Coevolutionary Framework for Managing Disease-Suppressive Soils. Annual Review of Phytopathology. 2011;49(1):47–67; doi: 10.1146/annurev-phyto-072910-095232.
Schlatter D, Kinkel L, Thomashow L, Weller D, Paulitz T. Disease suppressive soils: new insights from the soil microbiome. Phytopathology. 2017;107(11):1284–97.
Weller DM, Raaijmakers JM, Gardener BBM, Thomashow LS. Microbial populations responsible for specific soil suppressiveness to plant pathogens. Annual review of phytopathology. 2002;40(1):309–48.
Landa BB, Mavrodi OV, Schroeder KL, Allende-Molar R, Weller DM. Enrichment and genotypic diversity of phlD-containing fluorescent Pseudomonas spp. in two soils after a century of wheat and flax monoculture. FEMS Microbiology Ecology. 2006;55(3):351–68.
Huber D. The description and occurrence of suppressive soils. Suppressive soils and plant disease. 1982:1–7.
Smiley RW. Colonization of wheat roots by Gaeumannomyces graminis inhibited by specific soils, microorganisms and ammonium-nitrogen. Soil Biology and Biochemistry. 1978;10(3):175–9.
Mazzola M. Assessment and management of soil microbial community structure for disease suppression. Annu Rev Phytopathol. 2004;42:35–59.
Mazzola M. Manipulation of rhizosphere bacterial communities to induce suppressive soils. Journal of nematology. 2007;39(3):213.
Smith M, Hartnett D, Rice C. Effects of long-term fungicide applications on microbial properties in tallgrass prairie soil. Soil Biology and Biochemistry. 2000;32(7):935–46.
Doherty JR, Botti-Marino M, Kerns JP, Ritchie DF, Roberts JA. Response of microbial populations on the creeping bentgrass phyllosphere to periodic fungicide applications. Plant Health Progress. 2017;18(2):44–9.
Wang X, Lu Z, Miller H, Liu J, Hou Z, Liang S, et al. Fungicide azoxystrobin induced changes on the soil microbiome. Applied Soil Ecology. 2020;145:103343.
Lloyd AW, Percival D, Yurgel SN. Effect of fungicide application on lowbush blueberries soil microbiome. Microorganisms. 2021;9(7):1366.
Noel ZA, Longley R, Benucci GMN, Trail F, Chilvers MI, Bonito G. Non-target impacts of fungicide disturbance on phyllosphere yeasts in conventional and no-till management. ISME communications. 2022;2(1):19.
Latin R. A practical guide to turfgrass fungicides, 2nd ed. American Phytopathological Society (APS Press); 2021.
Salgado-Salazar C, Beirn LA, Ismaiel A, Boehm MJ, Carbone I, Putman AI, et al. Clarireedia: A new fungal genus comprising four pathogenic species responsible for dollar spot disease of turfgrass. Fungal biology. 2018;122(8):761–73.
Tredway LP, Tomaso-Peterson M, Kerns JP, Clarke BB. Compendium of Turfgrass Diseases. APS Press; 2023.
Walsh B, Ikeda SS, Boland GJ. Biology and management of dollar spot (Sclerotinia homoeocarpa): An important disease of turfgrass. HortScience. 1999;34(1):14.
Sapkota S, Catching KE, Raymer PL, Martinez-Espinoza AD, Bahri BA. New approaches to an old problem: Dollar spot of turfgrass. Phytopathology®. 2022;112(3):469–80.
Sang H, Hulvey J, Popko Jr JT, Lopes J, Swaminathan A, Chang T, Jung G. A pleiotropic drug resistance transporter is involved in reduced sensitivity to multiple fungicide classes in S clerotinia homoeocarpa (FT B ennett). Molecular plant pathology. 2015;16(3):251–61.
Staff G: 2015 State of the industry report. In: Golf Course Industry. https://www.golfcourseindustry.com/article/gci0115-golf-state-industry-report-2015/: GIE Media, Inc; 2015.
Tomer V, Sangha JK, Ramya H. Pesticide: An appraisal on human health implications. Proceedings of the National Academy of Sciences, India Section B: Biological Sciences. 2015;85:451 – 63.
Pimentel D, Acquay H, Biltonen M, Rice P, Silva M, Nelson J, et al. Environmental and economic costs of pesticide use. BioScience. 1992;42(10):750–60.
Goodman D, Burpee L. Biological control of dollar spot disease of creeping bentgrass. Phytopathology. 1991;81(11):1438–46.
Rodriguez F, Pfender WF. Antibiosis and antagonism of Sclerotinia homoeocarpa and Drechslera poae by Pseudomonas fluorescens Pf-5 in vitro and in planta. Phytopathology. 1997;87(6):614–21.
Shim T-S, Jung W-C, Do K-S, Shim G-Y, Lee J-H, Choi K-H. Development of antagonistic microorganism for biological control of dollar spot of turfgrass. Asian Journal of Turfgrass Science. 2006;20(2):191–201.
Roberts JA, Doherty JR. Limitations and Adoption Strategies for Biological Management of Turfgrass Pathogens. 2023.
Koch P, Hockemeyer K, Buczkowski E. Evaluating biological and oil-based fungicides for dollar spot suppression on turfgrass. Agronomy Journal. 2021;113(5):3808–18.
Azeem M, Hale L, Montgomery J, Crowley D, McGiffen Jr ME. Biochar and compost effects on soil microbial communities and nitrogen induced respiration in turfgrass soils. Plos one. 2020;15(11):e0242209.
Stacey N, Lewis R, Davenport J, Sullivan T. Composted biosolids for golf course turfgrass management: impacts on the soil microbiome and nutrient cycling. Applied Soil Ecology. 2019;144:31–41.
Doherty JR, Roberts JA. Investigating chemical and biological control applications for Pythium root rot prevention and impacts on creeping bentgrass putting green rhizosphere bacterial communities. Plant Disease. 2022;106(2):641–7.
Chou M-Y, Shrestha S, Rioux R, Koch P, Druzhinina IS. Hyperlocal Variation in Soil Iron and the Rhizosphere Bacterial Community Determines Dollar Spot Development in Amenity Turfgrass. Applied and Environmental Microbiology. 2021;87(10):e00149-21; doi: doi:10.1128/AEM.00149-21.
Koch PL, Grau CR, Jo Y-K, Jung G. Thiophanate-methyl and propiconazole sensitivity in Sclerotinia homoeocarpa populations from golf courses in Wisconsin and Massachusetts. Plant disease. 2009;93(1):100–5.
Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nature methods. 2012;9(7):676–82.
Cox MS, Deblois CL, Suen G. Assessing the response of ruminal bacterial and fungal microbiota to whole-rumen contents exchange in dairy cows. Frontiers in Microbiology. 2021;12:665776.
Lundberg DS, Yourstone S, Mieczkowski P, Jones CD, Dangl JL. Practical innovations for high-throughput amplicon sequencing. Nature methods. 2013;10(10):999–1002.
Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nature methods. 2016;13(7):581–3.
Abarenkov K, Nilsson RH, Larsson K-H, Alexander IJ, Eberhardt U, Erland S, et al. The UNITE database for molecular identification of fungi–recent updates and future perspectives. The New Phytologist. 2010;186(2):281–5.
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic acids research. 2012;41(D1):D590-D6.
Dixon P. VEGAN, a package of R functions for community ecology. Journal of Vegetation Science. 2003;14(6):927–30.
McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PloS one. 2013;8(4):e61217.
Costa PBd, Benucci GMN, Chou M-Y, Wyk JV, Chretien M, Bonito G, Turgeon BG. Soil Origin and Plant Genotype Modulate Switchgrass Aboveground Productivity and Root Microbiome Assembly. mBio. 2022;13(2):e00079-22; doi: doi:10.1128/mbio.00079-22.
Liaw A, Wiener M. Classification and regression by randomForest. R news. 2002;2(3):18–22.
Kursa MB, Rudnicki WR. Feature selection with the Boruta package. Journal of statistical software. 2010;36:1–13.
Wehrens R, Mevik B-H. The pls package: principal component and partial least squares regression in R. 2007.
Mendes R, Kruijt M, De Bruijn I, Dekkers E, Van Der Voort M, Schneider JH, et al. Deciphering the rhizosphere microbiome for disease-suppressive bacteria. Science. 2011;332(6033):1097–100.
Jiang G, Zhang Y, Gan G, Li W, Wan W, Jiang Y, et al. Exploring rhizo-microbiome transplants as a tool for protective plant-microbiome manipulation. ISME Communications. 2022;2(1):10.
Wei Z, Gu Y, Friman V-P, Kowalchuk GA, Xu Y, Shen Q, Jousset A. Initial soil microbiome composition and functioning predetermine future plant health. Science advances. 2019;5(9):eaaw0759.
Kong W-L, Li P-S, Wu X-Q, Wu T-Y, Sun X-R. Forest tree associated bacterial diffusible and volatile organic compounds against various phytopathogenic fungi. Microorganisms. 2020;8(4):590.
Yu L, Zi H, Zhu H, Liao Y, Xu X, Li X. Rhizosphere microbiome of forest trees is connected to their resistance to soil-borne pathogens. Plant and Soil. 2022;479(1):143–58.
Williams GM, Ginzel MD. Forest and plantation soil microbiomes differ in their capacity to suppress feedback between Geosmithia morbida and rhizosphere pathogens of Juglans nigra seedlings. Phytobiomes Journal. 2022;6(1):56–68.
Bailey K, Lazarovits G. Suppressing soil-borne diseases with residue management and organic amendments. Soil and tillage research. 2003;72(2):169–80.
Bonanomi G, Antignani V, Capodilupo M, Scala F. Identifying the characteristics of organic soil amendments that suppress soilborne plant diseases. Soil Biology and Biochemistry. 2010;42(2):136–44.
Boulter JI, Boland GJ, Trevors JT. Evaluation of Composts for Suppression of Dollar Spot (Sclerotinia homoeocarpa) of Turfgrass. Plant Disease. 2002;86(4):405–10; doi: 10.1094/pdis.2002.86.4.405.
Abdel-Azeem AM. Taxonomy and biodiversity of the genus Chaetomium in different habitats. In: Recent Developments on Genus Chaetomium. Springer; 2020. p. 3–77.
Soytong K, Kanokmedhakul S, Kukongviriyapa V, Isobe M. Application of Chaetomium species (Ketomium) as a new broad spectrum biological fungicide for plant disease control. Fungal Divers. 2001;7:1–15.
Wang M, Xue J, Ma J, Feng X, Ying H, Xu H. Streptomyces lydicus M01 regulates soil microbial community and alleviates foliar disease caused by Alternaria alternata on cucumbers. Frontiers in Microbiology. 2020;11:942.
Riviezzi B, Cagide C, Pereira A, Herrmann C, Lombide R, Lage M, et al. Improved nodulation and seed yield of soybean (Glycine max) with a new isoflavone-based inoculant of Bradyrhizobium elkanii. Rhizosphere. 2020;15:100219.
Nguyen HP, Ratu STN, Yasuda M, Teaumroong N, Okazaki S. Identification of Bradyrhizobium elkanii USDA61 Type III Effectors Determining Symbiosis with Vigna mungo. Genes. 2020;11(5):474.
Oliveira RS, Carvalho P, Marques G, Ferreira L, Pereira S, Nunes M, et al. Improved grain yield of cowpea (Vigna unguiculata) under water deficit after inoculation with Bradyrhizobium elkanii and Rhizophagus irregularis. Crop and Pasture Science. 2017;68(11):1052–9.
Zytynska SE, Eicher M, Rothballer M, Weisser WW. Microbial-Mediated Plant Growth Promotion and Pest Suppression Varies Under Climate Change. Frontiers in Plant Science. 2020;11; doi: 10.3389/fpls.2020.573578.
Wolf A, Fritze A, Hagemann M, Berg G. Stenotrophomonas rhizophila sp. nov., a novel plant-associated bacterium with antifungal properties. International journal of systematic and evolutionary microbiology. 2002;52(6):1937–44.
Raio A, Brilli F, Neri L, Baraldi R, Orlando F, Pugliesi C, et al. Stenotrophomonas rhizophila Ep2. 2 inhibits growth of Botrytis cinerea through the emission of volatile organic compounds, restricts leaf infection and primes defense genes. Frontiers in Plant Science. 2023;14:1235669.
Nguyen T-P, Meng D-R, Chang C-H, Su P-Y, Ou C-A, Hou P-F, et al. Antifungal mechanism of volatile compounds emitted by Actinomycetota Paenarthrobacter ureafaciens from a disease-suppressive soil on Saccharomyces cerevisiae. Msphere. 2023;8(5):e00324-23.
Zhang W, Krohn K, Draeger S, Schulz B. Bioactive isocoumarins isolated from the endophytic fungus Microdochium bolleyi. Journal of natural products. 2008;71(6):1078–81.
Shadmani L, Jamali S, Fatemi A. Biocontrol activity of endophytic fungus of barley, Microdochium bolleyi, against Gaeumannomyces graminis var. tritici. Mycologia Iranica. 2018;5(1):7–14.
Evueh G, Osemwegie O. Evaluation Phylloplane Fungi as Biocontrol Agent of Corynespora Leaf Disease of Rubber (Hevea brasiliensis Muell. ARG.). World Journal of Fungal and Plant Biology. 2011;2(1):01–5.
Tan Y, Cui Y, Li H, Kuang A, Li X, Wei Y, Ji X. Rhizospheric soil and root endogenous fungal diversity and composition in response to continuous Panax notoginseng cropping practices. Microbiological Research. 2017;194:10–9.
Komai S-i, Hosoe T, Itabashi T, Nozawa K, Yaguchi T, Fukushima K, Kawai K-i. New penicillide derivatives isolated from Penicillium simplicissimum. Journal of natural medicines. 2006;60:185–90.
Khokhar I, Mukhtar I, Mushtaq S. Antifungal effect of Penicillium metabolites against some fungi. Archives of Phytopathology and Plant Protection. 2011;44(14):1347–51.
Shimizu K, Hossain MM, Kato K, Kubota M, Hyakumachi M. Induction of defense responses in cucumber plants by using the cell-free filtrate of the plant growth-promoting fungus Penicillium simplicissimum GP17-2. Journal of Oleo Science. 2013;62(8):613–21.
Dodd S, Stewart A. Biological control of Pythium induced damping-off of beetroot (Beta vulgaris) in the glasshouse. New Zealand journal of crop and horticultural science. 1992;20(4):421–6.
Yuan Y, Feng H, Wang L, Li Z, Shi Y, Zhao L, et al. Potential of endophytic fungi isolated from cotton roots for biological control against verticillium wilt disease. PLoS one. 2017;12(1):e0170557.
Esmail SM, Draz IS, Saleem MH, Mumtaz S, Elsharkawy MM. Penicillium simplicissimum and Trichoderma asperellum counteract the challenge of Puccinia striiformis f. sp. tritici in wheat plants. Egyptian Journal of Biological Pest Control. 2022;32(1):1–9.
Pan F, El-Kashef DH, Kalscheuer R, Mueller WE, Lee J, Feldbruegge M, et al. Cladosins LO, new hybrid polyketides from the endophytic fungus Cladosporium sphaerospermum WBS017. European Journal of Medicinal Chemistry. 2020;191:112159.
Pan F, Yang N, Zhu X, Yu C, Jiang M, Jiang Y, et al. Discovery of a natural hybrid polyketide produced by endophytic cladosporium sphaerospermum for biocontrol of phytopathogenic fungus Botrytis cinerea. Journal of Agricultural and Food Chemistry. 2023;71(32):12190–202.
Lou J, Fu L, Luo R, Wang X, Luo H, Zhou L. Endophytic fungi from medicinal herb Salvia miltiorrhiza Bunge and their antimicrobial activity. Afr J Microbiol Res. 2013;7(4):5343–9.
Bright M, Bulgheresi S. A complex journey: transmission of microbial symbionts. Nature Reviews Microbiology. 2010;8(3):218–30.
Liu H, Brettell LE, Singh B. Linking the phyllosphere microbiome to plant health. Trends in Plant Science. 2020;25(9):841–4.
Tkacz A, Bestion E, Bo Z, Hortala M, Poole PS. Influence of plant fraction, soil, and plant species on microbiota: a multikingdom comparison. MBio. 2020;11(1):10.1128/mbio. 02785–19.
Vitoratos AG. Mode of action and genetic analysis of resistance to fluazinam in Ustilago maydis. Journal of Phytopathology. 2014;162(11–12):737–46.
Long JW, Siegel MR. Mechanism of action and fate of the fungicide chlorothalonil (2, 4, 5, 6-tetrachloroisophthalonitrile) in biological systems: 2. In vitro reactions. Chemico-biological interactions. 1975;10(6):383–94.
Niemi RM, Heiskanen I, Ahtiainen JH, Rahkonen A, Mäntykoski K, Welling L, et al. Microbial toxicity and impacts on soil enzyme activities of pesticides used in potato cultivation. Applied Soil Ecology. 2009;41(3):293–304.
Baćmaga M, Wyszkowska J, Kucharski J. The influence of chlorothalonil on the activity of soil microorganisms and enzymes. Ecotoxicology. 2018;27(9):1188–202.
Liao J, Luo L, Zhang L, Wang L, Shi X, Yang H, et al. Comparison of the effects of three fungicides on clubroot disease of tumorous stem mustard and soil bacterial community. Journal of Soils and Sediments. 2022:1–16.
Liu C, Yang Z, He P, Munir S, He P, Wu Y, et al. Fluazinam positively affected the microbial communities in clubroot cabbage rhizosphere. Scientia horticulturae. 2019;256:108519.
EPA US. Chlorothalonil: Reregistration eligibility decision. 1999.
Jiménez RR, Alvarado G, Ruepert C, Ballestero E, Sommer S. The fungicide chlorothalonil changes the amphibian skin microbiome: a potential factor disrupting a host disease-protective trait. Applied Microbiology. 2021;1(1):26–37.
Meena RS, Kumar S, Datta R, Lal R, Vijayakumar V, Brtnicky M, et al. Impact of agrochemicals on soil microbiota and management: A review. Land. 2020;9(2):34.
Thilakarathna MS, Raizada MN. A meta-analysis of the effectiveness of diverse rhizobia inoculants on soybean traits under field conditions. Soil Biology and Biochemistry. 2017;105:177–96.
Knežević M, Berić T, Buntić A, Delić D, Nikolić I, Stanković S, Stajković-Srbinović O. Potential of root nodule nonrhizobial endophytic bacteria for growth promotion of Lotus corniculatus L. and Dactylis glomerata L. Journal of Applied Microbiology. 2021;131(6):2929–40; doi: https://doi.org/10.1111/jam.15152.
Staley JT, Jenkins C, Konopka AE. Ancylobacter. Bergey's Manual of Systematics of Archaea and Bacteria. 2015:1–7.
Dedysh SN, Dunfield PF. Methylocella. Bergey's Manual of Systematics of Archaea and Bacteria. 2015:1–9.
Banik A, Mukhopadhaya SK, Dangar TK. Characterization of N 2-fixing plant growth promoting endophytic and epiphytic bacterial community of Indian cultivated and wild rice (Oryza spp.) genotypes. Planta. 2016;243:799–812.
Suarez C, Ratering S, Schäfer J, Schnell S. Ancylobacter pratisalsi sp. nov. with plant growth promotion abilities from the rhizosphere of Plantago winteri Wirtg. International Journal of Systematic and Evolutionary Microbiology. 2017;67(11):4500–6.
Sahin N, Gokler I, Tamer A. Isolation, characterization and numerical taxonomy of novel oxalate-oxidizing bacteria. Journal of Microbiology. 2002;40(2):109–18.
Lang E, Swiderski J, Stackebrandt E, Schumann P, Sproer C, Sahin N. Description of Ancylobacter oerskovii sp. nov. and two additional strains of Ancylobacter polymorphus. International journal of systematic and evolutionary microbiology. 2008;58(9):1997–2002.
Rioux RA, Stephens CM, Koch PL, Kabbage M, Kerns JP. Identification of a tractable model system and oxalic acid-dependent symptom development of the dollar spot pathogen Clarireedia jacksonii. Plant Pathology. 2021;70(3):722–34.
Bahri BA, Parvathaneni RK, Spratling WT, Saxena H, Sapkota S, Raymer PL, Martinez-Espinoza AD. Whole genome sequencing of Clarireedia aff. paspali reveals potential pathogenesis factors in Clarireedia species, causal agents of dollar spot in turfgrass. Frontiers in Genetics. 2023;13:1033437.

No competing interests reported.

FigS1AlfaDivShan.pdf
Figure S1. ASV Shannon diversity for bacterial (a) and fungal (b) communities in turfgrass grown with different sources of transplanted field microbiomes. The asterisks indicate significant mean separation derived from T-test: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.
FigS2HMSiBac.pdf
Figure S2. Heat maps showing the relative abundances of the bacterial orders of the microbiome transplanted turfgrass rhizosphere soil at Clarireediainoculation.
FigS3HMSiFun.pdf
Figure S3. Heat maps showing the relative abundances of the fungal orders of the microbiome transplanted turfgrass rhizosphere soil at Clarireediainoculation.
FigS4RFline2022.pdf
Figure S4. Random forest prediction lines by using the selected bacterial (a, b and c) and fungal (d, e and f) predictors to predict the turfgrass greenness (disease suppressiveness).
FigS5Corrdiseasenitrogen08102022.pdf
Figure S5. Correlation analysis of N application quantity in the field with the field microbiome transplanted turfgrass greenness, the indicator for dollar spot suppressiveness, after incubation with Clarirdeeia under disease favorable condition for 20 days.
FigS6plsModelIomprtance.pdf
Figure S6. Importance rank of major fungicide used in predicting dollar spot suppressiveness in the Partial Least Squares regression model.
SuppTablessubmitted.docx
Table S1. Bacterial and fungal ASV dispersion test of turfgrass phyllosphere and rhizosphere soil samples. Table S2. Non-significant pairs of paired-PERMANOVA analyses for bacterial and fungal communities of field microbiome transplanted turfgrass. The pairs not shown in this table all have adjusted p-value less than the significant threshold 0.05. P-values were adjusted for multiple comparison with Benjamini-Hochberg correction.

Fungicide use intensity influences the soil microbiome and fungal disease suppressiveness in amenity turfgrass

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

The presence of disease suppressive soil in turfgrass

Fungicide intensity impacts dollar spot suppressiveness

Potential dollar spot inhibitory and turf health promoting microbes in the soil

Potential dollar spot inhibitory and turf health promoting microbes in the phyllosphere

Untargeted effect of fungicide may modulate plant beneficial microbes

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1