impact of Bacillus subtilis QST713 mushroom grain spawn treatment on yield and green mould control

SUMMaRY A biofungicide based on Bacillus subtilis QST713 was tested for impact on yield and efficacy against a Trichoderma aggressivum f. europaeum T77 strain from Serbia by coating mushroom grain spawn and comparing the results with the chemical fungicide prochloraz manganese in a mushroom growing room. The tested B. subtilis QST713 strain did not inhibit mycelial growth of Agaricus bisporus in plots free of the pathogen, showing an impact on yield of 91.95%, which was not significantly different from an untreated control. As for the efficacy of the fungicides used against T. aggressivum f. europaeum T77, there were no significant differences between a prochloraz manganese casing treatment, and B. subtilis QST713 coating on mushroom grain spawn, as the efficacy was 70.37 and 53.09%, respectively. These results implied that the biofungicide based on B. subtilis could serve as a harmless alternative to synthetic fungicides in mushroom production, especially during serious compost green muold outbreaks caused by T. aggressivum . Furthermore, the biofungicide should be applied alone because an antagonistic reaction was detected between the fungicide prochloraz and B. subtilis QST713.


iNTRODUCTiON
Green mould of mushrooms remains the most devastating disease causing great losses in edible mushroom production. In serious outbreaks, the disease is characterised by fast-growing dark green colonies in the substrate that prevents the production of mushroom frutiting bodies (Seaby, 1996). Its causal agents belong to the soil-inhabiting genus Trichoderma (teleomorph Hypocrea, Ascomycota, Hypocreales). Of more than 200 analysed species, several have been found to belong to the widely cultivated beneficial micromycetes, plant symbionts or antagonist phytopathogenic fungi and antibiotic and enzyme producers (Druzhinina & Kubicek, 2005). Initially, green mould appeared simultaneously in 1985/86 in the British Isles (Doyle, 1991) and North America (Rinker, 1993), and spread rapidly to other parts of Europe and other continents (Geels, 1997;Hermosa et al., 1999;Mamoun et al., 2000;Hatvani et al., 2007;Romero-Arenas et al., 2009;Clift & Shamshad, 2009;Hatvani et al., 2012;Kosanović et al., 2013). Aggressive biotypes were discribed as Trichoderma aggressivum f. aggresivum in North America and T. aggressivum f. europaeum Samuels & W. Gams in Europe (Samuels et al., 2002), and they were phylogenetically closely related to T. harzianum Rifai and emerged through population adaptation to environmenal conditions (Kredics et al., 2010).
A regular method of pathogen control on mushroom farms worldwide is the application of various fungicides. Only a few fungicides have been officially recommended in mushroom production, prochloraz and metraphenone in Europe and worldwide, and chlorothalonil, thiabendazol and tiophanate-methyl in North America (Grogan & Gaze, 2000). Fungicides are usually applied to mushroom casing, but the introduced compost pathogen T. aggressivum has made disease control much more complicated. Benzimidazole and prochloraz were the first fungicides applied to compost, but they could not stop T. aggressivum (Abosriwil & Clancy, 2004). Fungicide application to grain spawn was found to provide better control of green mould colonization than compost treatment with the fungicide. It seems that mushroom grain spawn alone provides an available carbohydrate source from which compost colonization of green mould can proceed and be established (Grogan & Fletcher, 1993). However, Trichoderma growth was not supported in unspawned compost. After thorough investigation of mushroom grain spawn treated with benzimidazoles, benomyl and carbendazim, these fungicides received registration as spawn grain treatments in England and Pennsylvania, USA (Grogan et al., 1996). In addition, Rinker and Alm (1997a, b) found that green mould management is possible only by unifrom coating of spawn grains with a fungicide. However, its effectiveness may be reduced when supplements are mixed into the compost at spawning. Considering that only 1.15 g of fungicide is needed to treat the spawn for 1 tonne of compost, spawn treatment represents a more economical use of the fungicide, compared to compost treatment, which would require 70 g of product per tonne of compost. Over time, Trichoderma spp. have developed resistance to benzimidazoles (Romaine et al., 2005;Grogan, 2008). Also, fungicides such as benomyl, carbendazim and prochloraz have shown to be susceptible to microbial degradation in soils (Fletcher et al., 1980;Yarden et al., 1990). However, none of these fungicides is currently approved for spawn treatment. Registered fungicides are applied at a much later stage of crop cycle, on casing soil, when disease outbreaks are more likely.
As fungicides may also reduce the mycelial growth of A. bisporus, there must be a balance between the benefit from limitation of undesirable contaminants and a possible reduced vigor of a mushroom crop. Furthermore, many chemicals are no longer approved for use. In addition, there is an increased demand for reduced pesticide use and growers will increasingly have to depend on other measures to control outbreaks rather than resort to chemicals (Grogan, 2008). An alternative to chemical control of Trichoderma green mould is the application of microorganisms as biocontrol agents. Savoie et al. (2001) reported inhibitory effects of Bacillus species to T. aggressivum growth. Furthermore, Védie & Rousseau (2008) reported the effectiveness of lactonase-producing bacteria, Bacillus subtilis (Ehrenberg) Cohn, in preventing green mould disease, revealing data on the application rates but providing neither results of biofungicide efficacy in green mould control nor its impact on mushroom yield. The most studied Bacillus subtilis strain used as a biofungicide, the QST713, has been recently renamed as Bacillus velezensis (Pandin et al., 2018a,b). This biocontrol strain was chosen for tests of biological efficiency and efficacy against Serbia-originating T. aggressivum f. europaeum by coating mushroom grain spawn, in comparison with the fungicide prochloraz manganese in the mushroom growing room.

Fungal species and culture conditions
The pathogenic fungus used in this study was obtained from a culture collection of the Institute of Pesticides and Environmental Protection. The isolate Trichoderma aggressivum f. europaeum T77 was obtained from mushroom compost in the Serbian mushroom farm Lisovići in 2010. The fungus was identified based on morpho-physiological characteristics and ITS1/ ITS4 sequence analyses (Kosanović et al., 2013). Fungal cultures were maintained on potato dextrose agar (PDA) at 4°C. For inoculum preparation, agar discs of fungal isolates were grown on PDA for three days at 22°C. In preparation for in vivo trials, conidia from three-day old cultures of both fungi were flooded with 10 ml of sterile distilled water and Tween 20 (v/v 0.01%), followed by filtration through double layers of cheesecloth.

Antifungal agents
The biofungicide Serenade® WP (AgraQuest, Canada) based on Bacillus subtilis (B. subtilis QST 713 [5.13 · 10 10 CFU/g] 15.7%; other ingredients 84.2%) was tested as a potential antifungal agent for treatment of spawn grain of A. bisporus Sylvan A15 strain against the isolate of T. aggressivum f. eurpaeum T77 in the mushroom growing room. The biological efficacy and effectiveness of the biofungicide was evaluated by comparison with the commercial fungicide prochloraz manganese (Octave® WP, Bayer Crop Science, Germany, content of prochloraz manganese complex 50%; kaolin 35%; and other ingredients 15%) (Table 1).

Tests in mushroom growing room
Plastic bags sized 0.60 x 0.4 x 0.25 m (l x w x h) with 18 kg of compost (Uča & Co. Vranovo, Smederevo, Serbia) spawned with A.bisporus A15 (Sylvan, Hungária zRt) were incubated at 24 o C for 18 days. Compost surface of each bag was divided by a wooden barrier into two equal sections, making total area of each experimental plot to be 0.12 m 2 . Compost was cased with 50 mm layer of black peat casing soil Terahum (Treset d.o.o., Veliko Gradište, Serbia), disinfected with peracetic acid 0.02% (Peral-S 15%, Vetprom, Belgrade, Serbia), 90 ml per m 2 of casing, and incubated at 21 o C for 8 days (case-run), and after that air temeperature was reduced to 16 o C. The casing soil was drenched with the tested fungicide prochloraz manganese by split application on the fourth and twenty-third day after casing (after the first flush). Ten days after spawning, the tested plots were inoculated using conidial suspension of the pathogen at a concentration of 10 4 conidia/ml per m 2 of casing soil by pouring 125 ml in holes of 2.5 cm depth in the middle of the experiental plot surface (upper side). Spawn grain of A. bisporus A15 were coated using doses for one bag of mushroom compost of 18 kg: 252 g spawn grain A. bisporus, 15 g gypsum, 1.5 g of B. subtilis QST713. Spawn, gypsum and B. subtilis were thoroughly mixed before spawning. The fungicide prochloraz manganese was applied at the standard product application rate: 0.6 g of active ingredient (a.i.) in 1.8 l H 2 O per 1 m 2 of casing surface. The treatments were: (1) compost + A. bisporus spawn (control); (2) compost + A. bisporus spawn grain + biofugicide B. subtilis QST 713; (3) compost + A. bisporus spawn grain + T. aggressivum f. europaeum inoculum; (4) compost + A. bisporus spawn grain + biofugicide B. subilis QST 713 + T. aggressivum f. europaeum inoculum; (5) compost + A. bisporus spawn grain + prochloraz manganese; (6) compost + A.
Inoculated and uninoculated bags without fungicide treatments were used as controls. The plots were arranged in a completely randomized block design with four replicates per treatment. The fruiting bodies were hand-picked in three successive production flushes: the first from day 14 to 23 after casing, the second from day 24 to 34, and the third from day 35 to 45. The harvested mushrooms were weighed and divided into two groups based on visual observation, i.e. with and without symptoms of green mould disease. The effect of fungicides on mushroom productivity was evaluated as biological efficacy (BE), calculated as the ratio of fresh weight of total fruiting body yield and the weight of dry spawned substrate, and expressed as %: where Ic -disease incidence in the inoculated control; It -disease incidence in treated samples (Gea et al., 2010). Disease incidence was recorded as a percentage of fruiting bodies with symptoms compared with those without symptoms.

Statistical analyses
Data were examined using one-way analysis of variance (ANOVA), including comparisons of means by F-test. The test was used to compare the significance of differences in data on the average biological efficacy and effectiveness of different biofungicide/fungicide treatments against T. aggressivum f. europaeum T77 in the mushroom growing room. In all analyses, the level of significance was at least P<0.05 (Sokal & Rohlf, 1995). Statistical data analysis was performed using the software Statistica for Windows 6.0 (Stat Soft Italia, 1997).

ReSULTS aND DiSCUSSiON
Small brown spots of a few millimeters were found on A. bisporus fruiting bodies in plots inoculated with T. aggressivum f. europaeum T77 16 days after casing. Larger spots and necrotic lesions measuring a few centimeters were observed several days later. Small white colonies were noted on the casing surface 27 days after casing. A few days later, the colonies become emerald green after profuse sporulation.
The tested B. subtilis QST713 did not inhibit the growth of A. bisporus mycelia in plots without the pathogen, which is consistent with the results of experiments conducted by Milijašević-Marčić et al. (2017) in button mushroom, and Nagy et al. (2012) in oyster mushroom (Pleurotus ostreatus). Conversely, Kim et al. (2008) reported slightly inhibited mycelial growth of edible mushrooms, enoki (Flammulina velutipes), shiitake (Lentinula edodes) and oyster mushrooms in Korea, as did Kosanović et al. (2013) in A. bisporus, too.
The productivity of mushroom fruiting bodies was the highest in two controls, inoculated and uninoculated, and in the uninoculated plot treated with prochloraz manganese (Table 2). There were no significant differences between those plots and the rest of treatments, except the inoculated plots treated with B. subtilis QST713 which had the lowest yield in the experiment. Furthermore, there were no significant differences between the previous and all other treatments with B. subtilis QST713, and those inoculated and treated with prochloraz manganese. Both Kosanović et al. (2013) and Milijašević-Marčić et al. (2017) also reported the least yield from inoculated B. subtilis treatments, i.e. different strains applied on casing soil, QST713 and B-38, although all treatments had lower BE in these experiments, perhaps because of the lower A. bisporus inoculation (0.7%) in both previous experiments, whereas it was double in our trial, i.e. we used 1.4% mycelia of A. bisporus spawning in the compost. Milijašević-Marčić et al. (2017) noted the highest yield in uninoculated control and uninoculated prochloraz manganese treatment, which corresponded to the results in our experiment. We found the lowest yield in inoculated plots supplemented with mushroom grain spawn coated with B. subtilis QST713 which was consistent with findings reported by Kosanović et al. (2013), where a Bacillus strain was applied on casing soil. Milijašević-Marčić et al. (2017) also found the highest yield in the same treatment, i.e. plots treated with B. subtilis QST713 and applied to casing. Uninoculated plots treated with prochloraz manganese gave also the highest yields in all compared experiments.
As for the fungicide efficacy in controlling T. aggressivum f. europaeum T77, there were no significant differences between treatment with the fungicide prochloraz manganese, applied on casing, and B. subtilis QST713 coated on spawn grains of A. bisporus (Figure 1). Furthermore, there were no significant differences between the two treatments with B. subtilis QST713, with or without prochloraz manganese. However, prochloraz showed higher efficacy when it was applied alone than together with the biofungicide. There is a possible antagonistic reaction between the fungicide and the biofungicide in their simultaneous application. Milijašević-Marčić et al. (2017) found no differences between treatments with prochloraz and B. subtilis QST713 applied alone, which corresponds to our results. Our data showed 70.4% efficacy of prochloraz manganese. Similarly, Kosanović et al. (2013) reported 71.6% efficacy against T. harzianum, while Milijašević-Marčić et al. (2017) found a much lower efficacy of that fungicide (49.4%) against T. aggressivum. It is noteworthy that in the experiment of Kosanović et al. (2013), T. harzianum was a casing pathogen and the soil was inoculated six days after casing. In our trials, inoculation with T. aggressivum was done 10 days after casing, while in the investigation of Milijašević-Marčić et al. (2017), T. aggressivum was added on the surface of compost on the very same day of spawning, at least 20-24 days earlier than in the two previous experiments. These experiments have revealed that infestation of compost at spawnig is significantly more devastating than when spawn run compost is infested at casing time, which is consistent with the findings of Rinker and Alm (1997a, b). In our experiment, the efficacy of B. subtilis QST713 coating mushroom grain spawn against T. aggressivum was 53.1%, which was slightly higher than the data from Kosanović et al. (2013) with B. subtilis QST713 applied to casing soil against the casing pathogen T. harzianum (44%), and Milijašević-Marčić et al. (2017) when Bacillus subtilis strain was used against the compost pathogen T. aggressivum (48%). Kosanović et al. (2013) also detected antagonism between prochlroaz manganese and B. subtilis QST713, when they were applied together at 80% and 20% of the respective standard product application rates, which showed an efficacy of 29.2% against T. harzianum, while we found 43.21% efficacy with simultaneous application of both biofungicide and fungicide at full standard application rates.
It could be inferred that the fungicide prochloraz had satisfactory effectiveness when T. aggressivum inoculation occurred later, while the biofungicide showed similar efficacy as the chemical fungicide under an early presence of the pathogen. These results imply that the biofungicide based on B. subtilis could serve as a harmless alternative to synthetic fungicides in mushroom production, especially in serious compost green mould outbreaks caused by T. aggressivum. Also, the biofungicide should be applied alone because an antagonistic reaction was found between prochloraz and Bacillus. Certainly, similar experiments are required to confirm these findings and to investigate the efficacy of mushroom grain spawn coated with different strains of beneficial microorganisms against early pathogen infection.   Data are means of four replicates±SE, standard error of means; BE% -Biological efficiency = ratio of fresh weight of total mushroom yield and weight of dry spawned substrate; SEDs, standard error of differences=370.6; df, degrees of freedom=7; F=1.9; P-value=0.1.