Combination of yeast antagonists and Acibenzolar - S - Methyl reduced the severity of Fusarium head blight of wheat incited by Fusarium graminearum sensu stricto

The combination of yeast antagonists and Acibenzolar - S - Methyl (ASM) was tested against Fusarium graminearum on a spring wheat cultivar PAN3471. Two strains of Papiliotrema flavescens (Strains WL3 and WL6) and a strain of Pseudozyma sp . (MGO1) were combined with full strength ASM at anthesis, half strength ASM at anthesis and quarter strength ASM at late boot stages. The yeast and ASM treatments were applied prior to F . graminearum inoculation and disease progress was assessed over time. The combination of yeast and ASM treatments effectively reduced Fusarium Head Blight (FHB) severity and deoxynivalenol (DON) concentration compared to when the treatments were used alone. A positive correlation was observed between the Area Under Disease Progress Curve (AUDPC) and Percentage Seed Infection (PSI) (r = 0.44) whereas a negative correlation was observed between AUDPC and Hundred Seed Weight (HSW) (r = - 0.77) and PSI and HSW (r = -0.44). The best combination treatment providing the highest reduction in final disease severity (41.83%), high HSW and moderate PSI was 0.075 g/l ASM at anthesis plus P . flavescens strain WL3 . The highest DON reduction (19.35%) was by the treatment 0.075 g/l ASM at anthesis plus P . flavescens strain WL6. The best treatment was P . flavescens combined with 0.075 g/l ASM at anthesis. Although Pseudozyma sp. strain MGO1 did not provide the best FHB and DON reduction, its combination with ASM application im-proved disease control efficacy. To the best of our knowledge, this study presents the first report of the combination of P . flavescens and ASM in the management of FHB caused by F . graminearum in wheat plants.


Introduction
One of the major effects of Fusarium Head Blight (FHB) in crops is the production of mycotoxins in infected grains (1)(2)(3). These mycotoxins are a threat to human and animal health and have been reported to increase disease severity during infection by possibly disabling the plants' natural defence mechanisms (2,4,5). FHB infection is accompanied by the production of Fusarium-Damaged Kernels (FDKs) which cannot be used as either food, feed or seed (5,6). The mycotoxins that are produced by F. graminearum [teleomorph Gibberella zeae (Schwein.) Petch], a predominant causal agent of FHB, are deoxynivalenol (DON) (and its derivatives), nivalenol (NIV) (and its derivatives) and zearalenone (ZEA) (5,7,8). Apart from DON being the least harmful type B trichothecene mycotoxin, it is the most frequently detected mycotoxin and thus an indicator for mycotoxin contamination in FHB-infected grains (3,4,9). Previous studies indicate that pre-harvest control of FHB infection and development is the most promising means of reducing mycotoxin contamination on grains (5,6,10). This is because mycotoxin detoxification methods have limited efficacy on harvested grains and these methods have not yet been approved for use on grains with mycotoxin levels above acceptable limits (10).
Certain fungicides, such as triazole-based fungicides, have been used against FHB with reported efficacies (11)(12)(13)(14). In resistance breeding programmes, some progress has been made which includes the identification of possible sources of resistance (such as the Chinese cultivar 'Sumai 3') (2,5,15). Although some FHB control efforts have shown potential in disease reduction, there are currently no registered fungicides or bio-fungicides, and no commercially available resistant wheat varieties in most parts of the world (2,6,16). Current research into the control of FHB has been aimed at the use of natural antagonists, resistance breeding and integrated management (2,5,6,14,17). These methods are of research interest since they address the issues associated with fungicide use which include chemical residues in/on food and the development of resistant pathogen strains (due to excessive use) (2,4,18). Biological control of plant diseases has been studied over the years with reported efficacy. Its advantages include reduced environmental hazards (compared to chemical use), reduced likelihood of resistance development, and the production of durable plant protection (19). BCAs can be applied on plant residues (13), plant tissue (13), soil (20,21) and/or seed. Soil treated with Trichoderma harzianum and T. viride increased shoot dry weight, root dry weight and grain yield in the control of Sclerotium rolfsii (22). On the same study, the two Trichoderma species were reported to promote plant health by normalizing peroxidase (POX), phenylalanine ammonia lyase (PAL) and catalase (CAT) post inoculation with S. rolfsii. Streptomyces sp. RC 87B reduced FHB severity and DON by up to 39% and 85% respectively, on wheat during field trials (13). When applied on wheat stubble, Strptomyces sp. RC 87B reduced F. graminearum inoculum by at least 46% 90 days post inoculation (13).
It has been suggested that the best way to manage FHB is through integrated control strategies (11,15,23). Several studies on the incorporation of biological control agents (BCAs) in an integrated strategy for the control of FHB have been reported (24,25). The co-culture of Cryptococcus flavescens OH 182.9 and C. aureus OH 71.4 significantly reduced FHB severity (by 32% on average) compared to individual applications (25). The integration of resistance inducers with BCAs in the management of FHB in wheat has been previously studied (26). However, this study reports for the first time the combination of Acibenzolar-S-Methyl (ASM) and yeast antagonists in an integrated management strategy of FHB caused by F. graminearum. The aim of this study was to test the efficacy of combining ASM with yeast antagonists for the reduction of FHB severity and DON contamination in wheat. The effectiveness of the combined treatments as against each of the treatments alone was measured using the following parameters: (i) measure disease severity, (ii) Hundred Seed Weight (HSW), (iii) Percentage Seed Infection (PSI) and (iv) mycotoxin concentration (DON and ZEA) in harvested grains.

Planting and experimental design
Sixty-five planting pots of 25 cm diameter were filled up to 90% capacity with composted pine bark potting medium. Thereafter, spring wheat five seeds [cultivar PAN3471 obtained from Pannar Seed (Pvt) Ltd, Greytown, Republic of South Africa] were sown at even spacing in each pot which constituted an experimental unit. The trial consisted of 16 treatments (Table 1) with five replicates each. A completely randomised design was used for this experiment. The pots were placed in a growing area with insect netting (approximately 15% shading) and a drip irrigation system was used where each pot received water for 2 mins four times a day. Osmocote Exact Mini 5-6 M 15-3.9-9.1 + 1.2 Mg + TE [supplied by Greenhouse products (Pvt) Ltd, Helderkruin, Republic of South Africa (RSA)], an ammonium based slow-release fertilizer, was applied in each pot at a rate of 2.5 g/l of potting media.

Inoculum preparation
A F. graminearum (strain F20) conidia suspension previously stored at -80 o C, was thawed under a laminar flow cabinet at ambient temperature. This strain was obtained from the Discipline of Plant Pathology stock culture laboratory, University of KwaZulu-Natal, Pietermaritzburg, South Africa. The F. graminearum strain was previously isolated from infected wheat heads in a wheat cultivation field. Conidia were then streaked out onto fresh potato dextrose agar (PDA) plates and incubated at 25 o C for 5 days. Thereafter, the culture was subcultured by cutting out a 1 mm 3 agar plug from the actively growing edges of the mycelia and then placed faced down at the centre of a fresh PDA plate. This was repeated on 20 PDA plates and the plates were incubated at 25 o C for 7 days. Thereafter, the plates were placed under ultraviolet-A (UVA) light (360 nm wavelength) for 14 days to induce fungal sporulation.
The yeast strains Papiliotrema flavescens [strains WL3 and WL6, previously isolated from wheat (Triticum aestivum L.) leaves] and Pseudozyma sp. [strain MGO1 previously isolated from Mondo grass (Ophiopogon japonicus (L.f.) Ker-Gaw) leaves] with proven efficacy against F. graminearum in vitro (27) were used in this study. These strains had been identified by Inqaba Biotechnological industries (Pvt) Ltd (Muckleneuk, Pretoria, RSA) using Internal Transcribed Spacers (ITS) sequencing and molecular identification. The yeast strains were streaked out from their respective stock solutions (previously stored at -80 o C) onto fresh PDA plates with 10 replicates each and thereafter incubated at 25 o C for 5 days. Thereafter, 4 ml of sterile distilled water was pipetted onto each plate using a micropipette under aseptic conditions. Using a flamesterilized L-bent glass rod, the culture was suspended in the water by lightly rubbing the surface of the plate. The aliquot was decanted into a sterile and appropriately labelled conical flask. This was repeated for all the plates resulting in three flasks containing each yeast isolate.
Conidial suspensions of F. graminearum F 20 were prepared as above. The aliquot was transferred into a sterile Schott bottle which was vigorously shaken to allow the suspension of conidia in the solution. The aliquot was sieved through a sterile cheesecloth to remove mycelia and agar debris. Thereafter, the conidial concentration was adjusted to 1×10 5 conidia/ml using a haemocytometer and then made up to 10 l. The spore concentrations of each of the 3 yeasts were adjusted to 1×10 7 spores/ml and the solutions were made up to 4 l each.

Treatments application
Acibenzolar-S-Methyl (ASM) granules were purchased from Syngenta (Pvt) Ltd, Halfway house, Johannesburg, RSA. To prepare ASM concentrations, beakers were filled with tap water and placed on a bench top for an hour to allow the release of excess chlorine. ASM granules were weighed (0.019 g, 0.0563 g and 0.075 g) and each amount separately dissolved in 1 L of the tap water. The ASM solutions were transferred to previously cleaned and appropriately labelled 1 l pump spray bottles. With the nozzle adjusted to emit a fine mist, the plants were sprayed with the appropri-ate ASM solutions until runoff at the appropriate growth stages (Table 1).
Wheat heads were sprayed with the appropriate yeast spore suspensions until runoff according to the assigned treatments presented in Table 1. In all in vivo inoculations, plants of the same treatment were sprayed separately, away from the other plants to prevent spray drift. The heads were then covered with perforated, light-weight plastic bags for 24 hrs. to encourage humidity. Forty-eight hrs after yeast inoculation, the wheat heads were sprayed with conidial suspensions of F. graminearum until runoff and thereafter covered with the same plastic bags for 24 hrs to encourage disease development. Yeast and F. graminearum inoculations were each performed once. Disease severity was measured using a visual scale originally described by (29) and disease ratings were recorded in intervals over time. The experiment was repeated once.
When the plants had a golden-brown appearance and had reached maturity, wheat heads were cut off from the straws and put in appropriately labelled collection bags according to treatment replicates. Harvested grains were placed in a ventilating oven set at 55 o C for a period of 4 days. During this period, the bags were constantly monitored and shuffled to prevent heat damage of the grains. Thereafter, the wheat heads were threshed, and the seeds were transferred into appropriately labelled envelopes. These were stored in a cold room set at 4 o C for further experiments.
HSW and PSI were determined per treatment replicate for the 2 experiments. For the PSI, the seeds were surface sterilized, cultured on freshly prepared PDA plates and incubated at 25 o C for 4 days. Since each treatment had 5 replicates, each replicate had 3 plates which each had 15 seeds. The experiment was repeated once resulting in 30 plates per treatment. The number of Fusarium-infected seeds per plate was recorded and used to calculate the PSI using the following formula:

Mycotoxin analysis
The target mycotoxins were DON and ZEA since they are the most prevalent mycotoxins in FHB infections. The roQ TM QuEChERS kits KSO-8909 and KSO-9507 were used for sample extraction and dispersive Solid Phase Extraction (dSPE) respectively. These were purchased from Separations (Pvt) Ltd, Johannesburg, RSA. Mycotoxin extraction was performed according to (30), with modifications. Wheat seeds from the 2 experiments were pooled according to treatments for mycotoxin analysis. For each sample, the seeds were ground into fine powder using a Mikro-Feinmuhle-Cullati (MFC) plant grinder in the Plant Pathology seeds laboratory. A 5 g subsample was added into a 50 ml roQ QuEChERS extraction tube along with the following reagents: Milli-Q water (10 ml), acetonitrile with 5% formic acid (10 ml) and the contents of the roQ QuEChERS extraction packet (KSO-8909) which consisted of 4.0g MgSO4, 1.0 g NaCl, 1.0 g SCTD and 0.5 g SCDS (30).
The tube was shaken for 1 min by hand and then centrifuged at 4000 rpm (3000 g) for 5 mins (Beckman Coulter ® , Avanti ® J-26 XPI centrigufe) (30). Six ml (6 ml) of the supernatant were transferred into a roQ QuEChERS 15 ml centrifuge tube (KSO-9507) containing 900 mg MgSO4 and 150 mg primary secondary amine (PSA) (30). The tube was shaken by hand for 30 secs and then centrifuged as above (30). Thereafter, 1 ml of the supernatant was filtered through a 0.45 µm pore filter and transferred into a 1. The mobile phase consisted of aqueous 5 Mm ammonium acetate with 0.1% acetic acid, 5 mM ammonium acetate in methanol with 0.1% acetic acid, acetonitrile and Milli-Q water. The flow rate was 0.20 ml/min and the injection volume was 50 µl. The retention times for DON and ZEA were 2.563 and 10.193 mins respectively. Quantification was relative to external standards of 1-8 µg/ml in acetonitrile. Three quantification readings were conducted per sample.

Data analysis
HSW, PSI and disease severity data obtained were checked for homogeneity within the repeated trials and the data were thereafter pooled. Disease severity data was used to calculate the Area Under the Disease Progress Curve (AUDPC) for all treatments (31) before subjected to ANOVA. If the ANOVA was significant (P ≤ 0.05), the means were separated using the Duncan's Multiple Range Test (DMRT) at 5% significance level using SAS software Version 9.4 (32). Pairwise correlations were determined between AUDPC, HSW and PSI for the pooled data using the Spearman's correlation test (32). The rate of disease progress (r) was calculated using the Vanderplanks' logistic equation (33) expressed below: where; t1 = initial day of rating; t1 = final day of rating; x1 = initial disease value; x2 = final disease value.

Disease severity and seed infection studies
The 0.075 g/l ASM treatment at anthesis plus P. flavescens WL6 had the lowest disease severity rating in all rating days and thus the lowest final average disease severity (50.92%) compared to the control (87.53%) ( Table 2). This means that the number of infected spikes for the treatment 0.075 g/l ASM at anthesis plus P. flavescens WL6 were significantly less than those for the control treatment. The control treatment had the highest average disease severity in all rating days ( Table 2).
There were significant differences between the treatments for the AUDPC, HSW and PSI at p ≤ 0.0006, p ≤ 0.0001 and p ≤ 0.06 respectively ( Table 3). The lowest AUDPC and PSI values were observed for treatments 0.075 g/l ASM at anthesis plus P. flavescens WL6 and 0.075 g/l ASM at anthesis plus Pseudozyma sp. MGO1 respectively. The highest HSW was observed for the treatment 0.0563 g/l ASM at anthesis plus P. flavescens WL3. The PSI values of 8 out of 15 treatments were not significantly different from the control ( Table 3). As a result, some treatments with low AUDPC and HSW values were associated with high PSI values. An example of this was 0.075 g/l ASM at anthesis plus P. flavescens WL6, which had the lowest AUDPC units, the third highest HSW and a below-average PSI that was not significantly different from the control. The control had the highest AUDPC units, lowest HSW and highest PSI. The highest rate of disease progress (r) was observed for Pseudozyma sp. MGO1 and the lowest for 0.0563 g/l ASM at anthesis (Table 3). However, there was not much difference in the rate of disease progress between treatments and thus statistical analysis was omitted.

Correlation between AUDPC, HSW and PSI
Significant correlations were observed for all pairwise combinations (Table 4). A moderate positive correlation was observed between AUDPC and PSI (r = 0.44) which was sig-nificant at p = 0.0002. A strong and moderate negative correlation was observed between AUDPC and HSW (r = -0.77) and PSI and HSW (r = -0.44) respectively. These were significant at p < 0.0001 and p = 0.0003, respectively. The highest negative correlation was observed between AUDPC and HSW.

Mycotoxin analysis
A reduction of up to 19.45% in DON concentration was obtained and this was by the treatment 0.075 g/l ASM at anthesis plus P. flavescens WL3 ( Table 5). Although statistically similar to the DON concentration of the control, the other treatments with low DON concentrations were 0.075 g/l ASM at anthesis and 0.075 g/l ASM at anthesis plus P. flavescens WL6 respectively. The treatment 0.019 g/l ASM at late boot stage was the only treatment that had a DON concentration higher than that of the control. ZEA was not detected in all tested treatments and the control, as indicated Values followed by the same superscript letter are statistically identical  Table 5.

Discussion
In this study, we demonstrated that the combination of a plant defence inducer (ASM) with yeast antagoist reduced the severity of FHB and DON concentration in wheat. There is limited research on the integration of plant defence inducers with yeast antagonists in the management of F. graminearum in wheat. However, our results reveal good potential for FHB and DON reduction in wheat. When ASM was applied alone, the treatment 0.075 g/l ASM at anthesis had the lowest AUDPC units (1113.27), highest HSW (2.96 g) and the lowest DON concentration (12.45 ppm). When the biocontrol agents were applied alone, the treatment P. flavescens WL6 had the lowest AUDPC units (1158.57) and highest HSW (2.92 g), but P. flavescens WL3 had the lowest DON concentration (13.28 ppm). These treatments were previously tested prior to the current experiment (27) where their potential was studied. Although there is not much literature on the study of ASM for FHB control, ASM has been tested on other plant pathogens (34,35). In a study to test the effect of ASM against Botrytis cinera on table grapes, ASM reduced the incidence of gray mold by up to 85% (34). Yeast biocontrol agents such as Cryptococcus flavescens OH 182.9 have been studied and reported to effectively reduce FHB and DON concentrations in wheat (24)(25)(26).
The integration of ASM and biocontrol agents improved FHB reduction compared to when the treatments were applied alone. The best integration treatments providing the highest FHB and DON reduction were 0.075 g/ l ASM at anthesis plus P. flavescens WL6 (41.83%) and 0.075 g/l ASM at anthesis plus P. flavescens WL3 (19.35%) respectively. Moreover, the treatment 0.075 g/l ASM at anthesis plus P. flavescens WL6 had the highest reduction in AUDPC units (52.91%), high DON reduction (12.54%), high HSW (3.32 g) and a PSI below 50%. It is important to note that the best integration treatments were comprised of treatments that performed best amongst those applied alone, which was expected. In another study, the combination of 4 resistance inducers with the yeast antagonist Cryptococcus flavescens OH 182.9 did not significantly reduce FHB severity compared to when applied alone (24). However, lowest FHB severity values were often associated with integrated treatments (24). Our study, therefore, is the first to report effective reduction of FHB and DON in wheat following the integration of ASM treatments and yeast biocontrol agents.
An increase in FHB severity is accompanied by an increase in Fusarium-infected kernels and a reduction in seed weight (4,36). Similar correlations were also observed in our study thus aiding to the efficacy of these treatments in FHB management. Although majority the PSI values in our study were not significantly different to that of the control, the PSI of the tested treatments could potentially decrease with higher treatment application doses or integration with postharvest control methods. Pseudozyma sp. MGO1, which was isolated from the weed plant Ophiopogon japonicus (Mondo grass), although did not provide the best FHB and DON reduction compared to P. flavescens WL3 and P. flavescens WL6 treatments, the combination with ASM application resulted in increased efficacy. The treatment 0.075 g/l ASM at anthesis plus Pseudozyma sp. MGO1 was the best treatment amongst those treated with Pseudozy-  (27) contains in vitro and in vivo screening experiments on the antifungal effect of yeast isolates (such as MGO1) isolated from O. japonicus against F. graminearum. Therefore, this is novel work which shows potential and requires more research.
Although the two P. flavescens strains belong to the same species, the differences in their efficacies against F. graminearum shows that they could be different strains. Nevertheless, the combinations of the P. flavescens WL6 with 0.075 g/l ASM at anthesis was the overall best treatment in this study. Moreover, further research on the determination of the best inoculum dosage, frequency of application is required and could help improve the efficacy and reliability of the P. flavescens strains. Other studies in the control of plant diseases include testing ASM concentrations higher than the ones used in this study (34,35). Therefore, future research can be aimed at determining the efficacy of higher ASM concentrations in the control of FHB of wheat as well as the physiological effects of the treatments on the wheat plant.

Conclusion
This study provides extensive research into the efficacy of Acibenzolar-S-Methyl (ASM) in the integrated control of Fusarium Head Blight (FHB) incited by F. graminearum in wheat plants. The combination of ASM and the P. flavescens strains provided the best FHB and deoxynivalenol (DON) reduction compared to when either were applied alone. The highest reduction in final FHB severity and DON concentration was observed where 0.075g/l ASM was applied at anthesis in combination with P. flavescens strains WL6 and WL3 respectively. To the best of our knowledge, this study presents the first report of P. flavescens strains as combination treatments with ASM in the management of FHB caused by F. graminearum in wheat plants. Field studies are essential to determine the efficacy of combined use of ASM and the P. flavescens strains in environments similar to those present in commercial wheat cultivation systems.