Characterization of Sarocladium oryzae and its reduction potential of rice leaf blast 1

cause grain yield losses of up to 100 % (Prabhu

PALAVRAS-CHAVE: Magnaporthe oryzae; cerulenina; biocontrole; atividade enzimática.e-ISSN 1983-4063 -www.agro.ufg.br/pat-Pesq.Agropec.Trop., Goiânia, v. 47, n. 1, p. 41-52, Jan./Mar.2017 widely used method, however, it has negative impacts on the environment (Meng et al. 2015), besides promoting the resistance of certain pathogen molecules, resulting in the effectiveness loss of some active ingredients (Oliveira et al. 2015).One of the most efficient rice leaf blast control methods is the use of resistant cultivars.However, this resistance is rapidly overcome in the field, due to the high variability and complexity of the disease-causing pathogen (Chen et al. 2013).
Recently, biological control agents have been suggested as an additional management strategy (Nie et al. 2014).These agents can act as direct antagonists, inhibiting the growth and/or development of plant pathogens, or resistance inducers, promoting a resistance to biotic and/or abiotic stresses by activating gene expression and metabolic pathways, thereby triggering defense mechanisms such as systemic acquired resistance and induced systemic resistance (Van Loon et al. 1998, Shoresh et al. 2010).
The systemic acquired resistance is normally triggered by localized infection and provides longterm systemic resistance.It involves the activation of pathogenesis-related genes with antimicrobial activity which encode proteins that use salicylic acid as a signaling molecule (Durrant & Dong 2004, Pieterse et al. 2009), with salicylic acid levels increasing when the pathogen is detected (Mishina & Zeier 2007, Tsuda et al. 2008).The induced systemic resistance occurs as a response to root colonization by nonpathogenic agents that can phenotypically induce this type of resistance (Van Loon et al. 1998, Pieterse et al. 2009), and requires additional signaling components not dependent on salicylic acid (Shoresh et al. 2010).Both systemic acquired resistance and induced systemic resistance have been employed successfully in rice, using different biotic and abiotic resistance inducers (Smith & Métraux 1991, Manandhar et al. 1998, Tsukamoto et al. 1999, Ashizawa et al. 2005, Filippi et al. 2007and 2011, Sena et al. 2013).
A number of studies have been conducted in recent years using the rice blast pathosystem and different biocontrol agents, such as rhizobacteria, Epicoccum sp. and Cladosporium sp.(Fillipi et al. 2011, Sena et. al. 2013, Chaibub et al. 2016), which can act as resistance inducers.The Sarocladium oryzae [(Sawada) W. Gams & D. Hawksw] fungus is the causal agent of sheath rot, however, in rice, this disease only occurs in the stem during the reproductive phase and is not found in leaves.
This study aimed at evaluating the variability, in relation to cerulenin production, of 28 S. oryzae isolates, as well as the S. oryzae potential to reduce the severity of rice leaf blast, in addition to quantifying the activity of enzymes linked to plant defense mechanisms.

MATERIAL AND METHODS
The experiments were conducted at the Embrapa Arroz e Feijão, in Santo Antônio de Goiás, Goiás State, Brazil, from 2013 to 2014, using 32 isolates from its collection of multifunctional microorganisms and fungi.A total of 28 S. oryzae isolates, as well as one isolate of the pathogens Magnaporthe oryzae, Cochliobolus miyabeanus, Thanatephorus cucumeris and Monographella albescens, were used.
The antagonistic potential of the S. oryzae isolates against M. oryzae, C. miyabeanus, T. cucumeris and M. albescens fungi was assessed using the paired cultures method (Romeiro 2007).Mycelial discs with 5 mm in diameter were placed 45 mm apart on opposite sides of a Petri dish containing potato-dextrose-agar growth medium, with one disc consisting of the pathogen under study and the other a S. oryzae isolate.In the control treatment, pathogens were grown separately with a mycelial disc in the center of the dish.The inhibition zone was measured as soon as the control reached full growth.Analyses were carried out using a completely randomized design and the experiment was conducted in triplicate and means compared by the Scott-Knott test (p ≤ 0.05), using the SASM-Agri software (Canteri et al. 2001).All pathogens were used in the asexual phase.
For the cerulenin quantification, the S. oryzae isolates were grown separately on potato-dextroseagar medium and incubated at 25 ºC.After 10 days, a Characterization of Sarocladium oryzae and its reduction potential of rice leaf blast e-ISSN 1983-4063 -www.agro.ufg.br/pat-Pesq.Agropec.Trop., Goiânia, v. 47, n. 1, p. 41-52, Jan./Mar.2017 5 mm-wide mycelial disc was transferred to 100 mL of a liquid medium (Omura 1976) containing 1.0 % of glucose, 3.0 % of glycerol, 0.5 % of peptone and 0.2 % of sodium chloride, and incubated for 8 days, at 150 rpm and 25 ºC, to obtain the crude extract.The extract was then submitted to a vacuum with chloroform at a ratio of 1:1 (v/v), crystallized and diluted in 2 mL of ethanol, resulting in the concentrated extract or cerulenin (Côrtes et al. 2014), which was quantified by high performance liquid chromatography using a Perkin Elmer Flexar TM system (Bills et al. 2004), in a reversed-phase C18 column measuring 150 mm x 4.6 mm x 5 μm, with a fixed temperature of 40 ºC, where the mobile phase B (acetonitrile) was diluted in water (A).A commercial cerulenin standard (Sigma) was used, obtaining a standard curve from concentrations of 0.00125-2 µg µL -1 .Readings were performed using the Chromera software (Perkin Elmer) at a wavelength of 210 nm.Analyses were carried out using a completely randomized design in triplicate and means compared by the Scott-Knott test (p ≤ 0.05), using the SASM-Agri software (Canteri et al. 2001).
The filtrates of the S. oryzae BRM 6461 and BRM 6493 isolates were assessed in terms of their effect on conidial germination and appressorium formation in M. oryzae.The filtrate was obtained based on the methodology proposed by Côrtes et al. ( 2014) and the M. oryzae conidial suspension according to Filippi & Prabhu (2001), and adjusted for a concentration of 1 x 10 4 conidia mL -1 .The suspensions were placed on a glass slide and covered with artificial superhydrophobic surface material.The treatments were: 10 μL of BRM 6461 + 10 μL of M. oryzae conidial suspension; 10 μL of BRM 6493 + 10 μL of M. oryzae conidial suspension; and 10 μL of M. oryzae conidial suspension + 10 μL of sterile distilled water.The experiment was performed in a completely randomized design at triplicate and analyzed after 1, 3, 6, 12 and 20 h of incubation (room temperature) and the data and means compared by the Tukey test (p ≤ 0.05), using the SASM-Agri software (Canteri et al. 2001).Sixty M. oryzae conidia were observed using an epifluorescence microscope in 0.1 % calcofluor-white stain (Sigma) at 40X magnification.Conidia were considered germinated on formation of the germ tube and appressoria were deemed formed when there was a globular structure at the base of the germ tube (cell differentiation), responsible for penetrating the host.
Nitrogen in the form of ammonium sulfate [(NH4) 2 SO 4 + Fe and Bo] was applied to the topsoil at 14 and 19 days after planting (DAP).The S. oryzae BRM 6461 isolate in the form of concentrated filtrate and conidial suspension (3 x 10 5 conidia mL -1 ) was sprayed at 48 h prior to the inoculation with M. oryzae (BRM 31295 isolate), at a concentration of 3 x 10 5 conidia mL -1 , at 21 DAP.The M. oryzae inoculum was obtained as described by Filippi & Prabhu (2001).A randomized block design was used, with three replications, each consisting of one tray containing around 80 plants, and six treatments: T1 = control (water); T2 = spraying with S. oryzae filtrate; T3 = spraying with S. oryzae conidial suspension; T4 = spraying with S. oryzae filtrate at 48 h before the inoculation with M. oryzae; T5 = spraying with conidial suspension at 48 h before the inoculation with M. oryzae; and T6 = inoculation with M. oryzae.
Rice leaf blast severity was assessed in three treatments: plant inoculated with M. oryzae (control) and plants sprayed with S. oryzae conidial suspension and BRM 6461 filtrate, at 48 h before the inoculation with M. oryzae.Twenty plants per treatment were evaluated by using a grading scale with scores ranging from 0 to 9 (Leung et al. 1988).Four assessments were conducted, being the first at the onset of symptom emergence.The data were used to calculate the area under the disease-progress curve (Campbell & Madden 1990), and the reduction in disease severity was determined by using the McKinney index (Balardin et al. 1990).Analyses were carried out in a completely randomized design and the data submitted to analysis of variance and means compared by the Tukey test (p ≤ 0.05), using the SASM-Agri software (Canteri et al. 2001).
The samples were ground with liquid nitrogen and the protein extract obtained using a lysis buffer [(10 Mm of Tris-HCl; 150 mM of NaCl; 2 Mm of EDTA (ethylenediaminetetraacetic acid); 2 Mm of DTT (2-mercaptoethanol); 1 Mm of PMSF (phenylmethylsulfonyl fluoride), Leptin (10 mg mL -1 ) and aprotinin (10 mg mL -1 )].Total protein was quantified by the colorimetric method (Bradford 1976) and readings were performed in a spectrophotometer, at a wavelength of 595 nm, using the Gen5 Data Analyses software (Biotek).The standard curve for total proteins with bovine serum albumin was obtained with different concentrations (0-1 mg mL -1 ).
The activity of chitinase (EC 3.1.1.14)and β-1,3-glucanase (EC 3.2.1.6)were determined as described by Pan et al. (1991), with modifications, using colloidal chitin and soluble laminarin as substrate, respectively.The reducing sugars resulting from the reactions were quantified using the 3,5-dinitrosalicylic acid method.Readings performed were adjusted to 540 nm, with the Gen5 Data Analyses software.Lipoxygenase (EC 1.13.11.12) activity was assessed using the method described by Axelrod et al. (1981) and readings were performed on a spectrophotometer, at a wavelength of 234 nm.Peroxidase (EC 1.11.1.7)activity was established according to the method proposed by Keesey (1987) and modified by Côrtes et al. (2008), using a spectrophotometer adjusted to a wavelength of 405 nm and coupled with the WinSpec 2.3 software, in the kinetic mode.Phenylananine ammonia lyase (EC 4.3.1.5)activity was evaluated based on the method described by Alunni et al. (2003), through quantification of the trans-cinnamic acid generated by hydrolysis of the L-phenylalanine and readings on a spectrophotometer adjusted to 290 nm.The results were expressed as specific activity.Salicylic acid was analyzed in accordance with Yalpani et al. (1991) and quantified by high performance liquid chromatography with mobile phase composed of 23 % of methanol and 77 % 20 mM of acetate buffer (pH 5.0), in a C18 column maintained at a fixed temperature of 35 ºC.The ultraviolet detector was adjusted to 280 nm and coupled with the Chromera software (Perkin Elmer).The data were submitted to statistical analysis using the Statistical Package for the Social Sciences software and means compared by the Tukey test (p ≤ 0.05).

RESULTS AND DISCUSSION
Of the 28 S. oryzae isolates analyzed, only six exhibited antagonistic activity against at least one of the pathogens studied (Table 1).BRM 6493 was the best antagonist, producing the largest zones of inhibition for the four pathogens tested.
In order to explain the rate of antagonism between the phytopathogens and the S. Oryzae fungus, isolates of the species were submitted to specific cultivation conditions to determine the production capacity of the secondary metabolite cerulenin.This molecule has been described as the main secondary metabolite responsible for the in vivo and in vitro antagonistic action of S. oryzae against rice pathogens (Sakthivel & Gnanamanickam 1986, Gnanamanickam & Mew 1991, Prabhu et al. 2007, Côrtes et al. 2014).The results of the present study indicated that 11 of the 28 isolates analyzed showed no detectable levels of this metabolite.BRM 6461 stood out from the remaining isolates, achieving a cerulenin production of 296 μg mL -1 (Table 2), more than twice as high as the second best ceruleninproducing isolate.
However, a comparison between data for in vitro antagonism and cerulenin production capacity found that isolates such as BRM 6493, which did not exhibit detectable cerulenin levels, were considered superior antagonists to the different pathogens.Thus, it is believed that in vitro antagonism was not only due to cerulenin production by some of the S. oryzae isolates, but also secondary metabolites such as helvolic acid, also often related to in vitro antagonism (Tschen & Wen 1980, Tschen et al. 1997).Another hypothesis is the production of other hitherto disregarded secondary metabolites, which may influence antagonistic behavior between species.Thus, it can be inferred that cerulenin was not the only molecule responsible for antagonistic action against pathogens in the in vitro mycelial development of S. oryzae.This information raises the possibility that molecules such as helvolic acid and others not described here, likely produced on a smaller scale, Characterization of Sarocladium oryzae and its reduction potential of rice leaf blast e-ISSN 1983-4063 -www.agro.ufg.br/pat-Pesq.Agropec.Trop., Goiânia, v. 47, n. 1, p. 41-52, Jan./Mar.2017 are acting as inhibitors of M. oryzae cell growth, in vitro, under the conditions described.
In addition to the antagonistic effect on mycelial growth, the inhibition of the conidial germination and appressorium formation of phytopathogens are of significant interest when the aim is to select biocontrol agents or molecules, particularly in terms of rice leaf blast management.In this respect, the S. oryzae BRM 6461 and BRM 6493 isolates were tested against M. oryzae conidia, exhibiting a delay in germ tube emergence in the first hour and 89.5 % and 85 % formation of M. oryzae appressoria, respectively (Table 3, Figure 1).The results shown by BRM 6461 were similar to those found by Côrtes et al. (2014) and Ohtake et al. (1999), who associated the reduced M. oryzae appressorium formation with increased cerulenin levels.
It is important to underscore that the metabolite cerulenin acts in the inhibition of lipid synthesis, one of the main components needed for the germ tube and appressorium formation in M. oryzae (Ebbole 2007).However, the BRM 6493 isolate displayed a marked inhibitory effect on appressorium formation: it did not produce cerulenin.The method used to produce the filtrate analyzed favors the cerulenin production and therefore hampers the helvolic acid production, which is generated in insignificant quantities (Omura 1976).This strengthens the hypothesis that secondary metabolites other than cerulenin and helvolic acid are produced and may be responsible for the inhibitory effect.
Based on the previous results, the S. oryzae BRM 6461 isolate was selected to assess the action of S. oryzae in reducing the severity of rice leaf blast.The experiment demonstrated that the concentrated filtrate or conidial suspension applied at 48 h before the inoculation with M. oryzae promoted reductions * Means followed by the same letter do not differ statistically according to the Skott-Knott test (p ≤ 0.05).2).However, although this difference is statistically significant, it is not enough to be explained by the difference in the cerulenin concentration potentially present on the leaf surface at the moment of interaction with the pathogen.A far higher effect on disease reduction was expected in the presence of the concentrated filtrate, containing high cerulenin levels, than the treatment consisting of S. oryzae conidia isolated, together with their locally produced metabolites, possibly including cerulenin.

Sarocladium oryzae
In this respect, the hypothesis tested was whether both the S. oryzae filtrate and conidia, in addition to exhibiting antagonistic action against the pathogen, might be acting indirectly on the defense system of the rice plants through a process called induced resistance, promoting the expression of pathogenesis-related proteins.The pathogenesisrelated protein expression was assessed by measuring the activity of five representative enzymes, in addition to the phytohormone salicylic acid.There are reports of increased expression of these enzymes and salicylic acid in potentially beneficial interactions among M. oryzae, rice and other microorganisms, confirming the induced resistance phenomenon and helping to suppress rice leaf blast (Filippi et al. 2014, Chaibub et al. 2016).
When applied in the absence of M. oryzae, the S. oryzae filtrate increased the chitinase activity (48 h after application), β-1,3-glucanase (5 h and 48 h after application) and peroxidase (48 h after application) in rice leaves, while for the conidial suspension the highest values were observed for β-1,3-glucanase (5 h, 24 h and 48 h after application) and peroxidase (5 h and 48 h after application).With respect to the enzyme activity in rice leaves sprayed with S. oryzae at 48 h before the inoculation with M. oryzae, the highest activity was recorded for lipoxygenase at 5 h when sprayed with the filtrate and at 24 h and 72 h with the conidial suspension.Phenylalanine ammonia lyase showed activity at 5 h after the treatment with the filtrate and after 72 h with the conidial suspension.No differences were found in the expression of chitinase, ß-1,3-glucanase, peroxidase and salicylic acid, in relation to the controls (Figure 3).GT = germ tube; A = appressoria.* Means followed by the same letter do not differ statistically according to the Tukey test (p ≤ 0.05).ns non-significant.

Figure 2 .
Figure 2. Area under the disease progress curve, where: a is the plant inoculated with Magnaporthe oryzae alone and b and c the plants sprayed with Sarocladium oryzae conidial suspension and BRM 6461 filtrate, at 48 h before the inoculation with M. oryzae, respectively.* Means differ according to the Tukey test (p ≤ 0.05).

Table 3 .
Number of conidia and appressoria formed during the exposure to Sarocladium oryzae filtrates of the BRM 6461 and BRM 6493 isolates.Means followed by the same letter do not differ statistically according to the Skott-Knott test (p ≤ 0.05).nd: non-detectable levels. *