Antifungal activity of Musa phyllosphere Bacillus pumilus strain against Mycosphaerella fijiensis Mileidy Cruz-Martín, Mayra AcostaSuárez, Eilyn Mena, Berkis Roque, Tatiana Pichardo & Yelenys AlvaradoCapó Tropical Plant Pathology e-ISSN 1983-2052 Volume 42 Number 2 Trop. plant pathol. (2017) 42:121-125 DOI 10.1007/s40858-017-0139-3 1 23 Your article is protected by copyright and all rights are held exclusively by Sociedade Brasileira de Fitopatologia. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”. 1 23 Author's personal copy Trop. plant pathol. (2017) 42:121–125 DOI 10.1007/s40858-017-0139-3 SHORT COMMUNICATION Antifungal activity of Musa phyllosphere Bacillus pumilus strain against Mycosphaerella fijiensis Mileidy Cruz-Martín 1 & Mayra Acosta-Suárez 1 & Eilyn Mena 1 & Berkis Roque 1 & Tatiana Pichardo 1 & Yelenys Alvarado-Capó 1 Received: 3 October 2016 / Accepted: 6 February 2017 / Published online: 27 February 2017 # Sociedade Brasileira de Fitopatologia 2017 Abstract Black leaf streak disease (BLSD; also known as black Sigatoka) caused by Mycosphaerella fijiensis Morelet is one of the main diseases affecting banana trees worldwide. Associated microorganisms in the phylloplane have been previously studied for their control. An exhaustive understanding of biocontrol mechanisms will potentiate the biological control agent. Therefore, strain Bacillus pumilus CCIBP-C5, isolated from banana leaves surface was evaluated for in vitro antagonism against M. fijiensis and the effect of culture filtrate on fungal growth and disease progress was determined. Bacterial cell and its diffused metabolites inhibited M. fijiensis growth. The culture filtrate of B. pumilus affected the mycelial growth, with antifungal effect on hyphal tip, membrane permeability and integrity. Also, a reduction of number of BLSD necrotic lesions and disease progression were found after application of the B. pumilus CCIBP-5 culture filtrate in banana plants. Results indicate that CCIBP-C5 has potential for biological control of BLSD. Keywords Black Sigatoka . Biological control . Culture filtrate . Greenhouse . Mycelial damage Black leaf streak disease (BLSD) caused by Mycosphaerella fijiensis Morelet reduces plant photosynthetic area and produces premature ripening of banana fruits (Churchill 2011). Section Editor: Bernardo A. Halfeld-Vieira * Mileidy Cruz-Martín mileidy@ibp.co.cu 1 Instituto de Biotecnología de las Plantas, Universidad Central “Marta Abreu” de Las Villas, Carretera a Camajuaní km 5.5, Santa Clara, Villa Clara, Cuba 54830 Chemical products have been widely used and have been highly successful during many years to control the disease. However, serious concerns on pesticide residues and environmental pollution have emerged recently. In addition, the development of fungicide resistance triggered the search for sustainable alternatives. Associated microorganisms of the phylloplane and the rhizosphere of multiple crops have been used for the control of various diseases. Ceballos et al. (2012) found that five percent of aerobic endospore forming bacteria from banana phyllosphere showed in vitro antifungal activity against M. fijiensis. Gutierrez-Monsalve et al. (2015) evaluated the effect of a microbial fungicide (MF) based on Bacillus subtilis EA-CB0015 against BLSD. The MF used in different programs with systemic fungicides reduced disease intensity up to 42.9%, and did not differ from the conventional program. The efficacy of MF depends mainly on the metabolites produced by the bacterial cells. However, the mechanism of biocontrol is still not fully understood. The current study is aimed at analyzing the antifungal effect of B. pumilus strain against M. fijiensis. For this purpose, the in vitro antifungal activity of bacterial cells and its culture filtrate was assayed. Finally, the effect of culture filtrate on artificially inoculated plants was determined. Bacillus pumilus CCIBP-C5 was isolated from banana phyllosphere and identified using the API 50 CHB kit (BioMérieux, Inc.). Mycosphaerella fijiensis strain CCIBPPf-83 was isolated from the leaves of the susceptible genotype ‘Grande naine’ (Musa AAA) and its identity was confirmed by PCR amplification of the internal transcribed spacer for ribosomal DNA, according to Johanson and Jeger (1993). The strain was stored in potato dextrose agar (PDA) culture medium at 4 °C until its subsequent use. M. fijiensis was grown in potato dextrose broth (PDB) medium for 15 days at 28 °C on a rotatory shaker at 120 rpm to obtain mycelial Author's personal copy 122 suspensions. The suspension was homogenized with UltraTurrax T25 for one minute and the concentration was determined using a hemocytometer. Both strains are kept in the Microbiological Culture Collection at Instituto de Biotecnología de las Plantas, Cuba. A dual culture assay was performed to explore the bacteria effect on M. fijiensis mycelia on PDA. An overnight grown culture of CCIBP-C5 was centrifuged (10 min at 4 °C and 12 000 × g) and cells were adjusted to an OD600 nm of 0.1 in deionized water. Seven microliters of bacterial suspension were inoculated in the center of plates containing PDA and fungal mycelial fragments (~5 x 105 fragments/mL). In the control plates, the bacterial suspension was replaced by sterile distilled water. The plates were kept at 28 °C for 72 h and inhibition zones were measured (mm). The experiment was repeated twice using two replicates. In order to determine the antifungal activities of diffused compounds, plates containing PDA and fungal mycelial fragments were covered with a cellophane membrane and an aliquot (3 μL) of bacterial suspension (OD600nm = 0.1) was placed in the middle of the plate. After incubation for 24 h at 28 °C, the membrane with the bacterial strain grown was removed, and the plate was incubated for 48 h at 28 °C. The fungal growth inhibition halo was measured and compared to control, where the bacterial suspension was replaced by sterile distilled water. The experiment was repeated three times. After that, CCIBP-C5 was grown for three days in nutrient broth (NB) medium at 30 °C and 120 rpm on an orbital shaker to obtain the culture filtrate (CF) containing the diffused compounds. Bacillus pumilus CF was obtained after centrifugation for 10 min at 4 °C and 12 000 × g followed by filtration through a 0.22 μm filter membrane. The CF was diluted (1:10) in M. fijiensis mycelial suspension and incubated for 48 h at 28 °C. The control treatment contained NB inoculated with mycelial suspension. Fungal growth was determined by absorbance reading according to Cruz-Martín et al. (2013) using 96 wells microtiter plates. In addition, mycelia were observed under the optical microscope (400x) and described. The experiment was repeated twice using eight replicates per treatment. The Congo red staining was used to allow visualization of chitin deposition at hyphal tip of M. fijiensis. After incubation with CF, Congo red was added to the fungal cultures to a final concentration of 1 mM. Ten minutes later the fluorescence was observed by confocal microscopy (excitation wavelength 543 nm; emission wavelength 560 to 635). The red staining of hyphal tip was an indication of growth inhibition. The effect of CF of CCIBP-C5 strain on fungal membrane integrity was assessed by determining the concentration of potassium in the growth medium. M. fijiensis mycelial suspensions grown after 48 h culture were inoculated into 2 mL 5 mM Tris-HCl (pH 7) buffer solution and adjusted to 5 x 105 fragments/mL. Next, the CF diluted in the buffer solution Trop. plant pathol. (2017) 42:121–125 (1:10) was applied and incubated for 48 h at 28 °C in darkness. Fresh NB (1:10) was used as control. The assay was performed in triplicate. The potassium concentration in the growth medium was determined by atomic absorption spectrophotometry (Unicam SP9) and results were expressed in mg/L. The effect of CF in oxidative stress was assessed by determination of lipid peroxidation and protein oxidation. The samples were obtained by inoculating the M. fijiensis suspension (~5 x 105 fragments/mL) in a flask with PDB medium and diluted CCIBP-C5 CF (1:10). The flasks were incubated for 48 h at 28 °C in the dark and mycelia were separated by centrifugation for 5 min at 4 °C and 10 000 × g. The extent of lipid peroxidation [malondialdehyde (MDA) as principal product] was estimated using the thiobarbituric acid assay described by Choi et al. (1996). Protein damage was determined by the advanced oxidation protein products (AOPP) assay (Matteuci et al. 2001) using Chloramine T as a standard. Micropropagated ‘Grande naine’ plants were used to evaluate the effect of CF in the plant. Plants were grown for three months in the greenhouse until they reached 20 cm of height and had more than three developed leaves. Preparation of M. fijiensis mycelia suspension, artificial inoculation and evaluation of symptom development and evolution were carried out following the protocol described by Leiva-Mora et al. (2010). The first three leaves of the plants were inoculated with the fungal suspension containing 1% gelatin. The CF (~3 mL per leaf) was applied on the abaxial surface of the leaf at 72 h after (CF A) and 72 h before (CF B) M. fijiensis inoculation. An equal number of plants inoculated with M. fijiensis and NB were used as controls. A complete randomized design with five plants per treatment was set up in greenhouse conditions (>80% humidity, 28 ± 2 °C) and the experiment was repeated twice. BLSD development was assayed every week during 70 days post inoculation (dpi) using the rating scale of Alvarado-Capó et al. (2003). The evaluation of inoculated leaves included: incubation time (defined as the time between inoculation and appearance of the first symptoms), disease development time (defined as number of days between inoculation and occurrence of lesions with whitish to gray centers that have a black border and are slightly depressed), area under disease progress curve (AUDPC) by Shanner and Finney (1977) and number of necrotic lesions. Furthermore, the efficacy of biocontrol was evaluated at 70 days according to Fu et al. (2010). In all experiments, the non-parametric KruskalWallis test was used for data analysis. Treatment means were separated by Mann-Whitney test using SPSS Statistics ver. 21.0 for Windows and the significance was set at P < 0.05. Bacillus pumilus CCIBP-C5 inhibited M. fijiensis growth (Fig. 1a). An inhibition zone (35.7 mm of radio) was observed at 72 h in the presence of the bacterial suspension. This corroborated the presence of bacteria of the genus Bacillus with Author's personal copy Trop. plant pathol. (2017) 42:121–125 123 Fig. 1 Antifungal activity of Bacillus pumilus CCIBP-C5 against Mycosphaerella fijiensis. a) Inhibition zone on mycelial growth by dual culture assay, b) Light micrographs of M. fijiensis hyphae after 48 h of incubation with nutrient broth culture medium (control) and after incubation with culture filtrate of CCIBP-C5 (CF) [note the swollen (black arrow) hyphal tip] c) Deformation and vacuolization of M. fijiensis mycelium in culture filtrate treatment d) Mycelial growth is arrested by diffused metabolite (yellow arrows indicate strong Congo red staining at the protuberances on the tips of hyphae) antifungal activity against M. fijiensis in the phyllosphere of Musa spp. that may constitute potential biological control agents. These results agree with those obtained by Ceballos et al. (2012). The antifungal activity was related to the production of diffused metabolites by the bacterial strain. This was confirmed by the effect of CF on M. fijiensis growth (Fig. 1b–d). The CF caused an inhibition of 45% of M. fijiensis growth in vitro with respect to control as measured by microtiter assay, suggesting an effect on the cellular structures of the fungus. Microscopical observations indicated that the untreated pathogen was completely filled, having intact fungal walls and defined membranes. However, it was observed that the CF of CCIBP-C5 produced morphological changes as shown by bulges or cracks in the hyphae tip of the mycelium as well as vacuolization (Fig. 1b–c). Zhou et al. (2008) detected similar deformations in mycelial morphology of Penicillium expansum (Link.) in the presence of Paenibacillus polymyxa culture filtrate and attributed it to peptidic metabolites probably acting on the cell membrane of the fungus and its permeability. Mycosphaerella fijiensis cultures treated with CF exhibited strong Congo red staining at the protuberances in the tips of the hyphae (Fig. 1d). These results suggest that CF arrests hyphal growth in fungal culture. Fungal growth inhibition is in accordance with the previous test of diffused metabolites of CCIBP-C5 bacterial cells with antifungal activity against M. fijiensis. Also, B. pumilus CCIBP-C5 produced metabolites that have an effect on M. fijiensis membrane. An oxidative damage at membrane level and an increase in its permeability were found. CF clearly stimulated an oxidative burst on the mycelium of M. fijiensis and the damage was observed at lipid and proteins levels. As shown in Fig. 2, a significant increase in MDA and AOPP concentrations was observed when CFCCIBP-C5 strain was applied. The results related to oxidative damage to lipids and proteins, main constituents of cell membranes, partly explain the deformations observed in M. fijiensis mycelium during exposure with the CF of CCIBP-C5 strain. Likewise, a significant increase of potassium ions in growth media was observed in the treated mycelium when compared to the control. Wang et al. (2002) noted that changes in the plasma membrane represent the first step in a cascade of events that promotes internal osmotic imbalance and cytoplasmic disorganization. This process is characterized by abundant vacuole formation, increased aggregation and loss of cytoplasmic organelles. The deformations observed in M. fijiensis hyphae in the presence of CF could be related to damages in the mycelium cell wall. These metabolites could be hydrolytic enzymes (chitinase and glucanase), that degrade chitin and β-1,3- Author's personal copy 124 Trop. plant pathol. (2017) 42:121–125 Fig. 2 Oxidative stress on Mycosphaerella fijiensis CCIBP-Pf-83 mycelia in the presence of bacterial culture filtrate at 48 h of incubation. a) Measurement of malondialdehyde (MDA) production as result of membrane lipoperoxidation (mycelium plus culture medium). b) Measurement of advanced oxidation protein products (AOPP) production. c) Measurement of concentrations of potassium in the growth medium. Differents letters above bars indicate significant differences by Kruskal-Wallis/ Mann-Whitney at p <0.05 (n = 3) glucan, constituents of ascomycetes cell wall. Likewise, Gutiérrez-Román et al. (2015) and Moreno et al. (2016) have proposed to reduce or complement chemical disease management programs against BLSD with the use of chitinolytic microorganisms. Also, in CF of B. pumilus CCIBP-C5, lipopeptides could be present. Ruangwong et al. (2012) found globular structures or swelling at the conidia ends of Colletotrichum gloeosporioides by the action of lipopeptides (iturins) present in the culture filtrate of B. subtilis. Bacillus lipopeptides have been studied for their antagonistic activity against a wide range of phytopathogenic microorganisms including viruses, bacteria, fungi and oomycetes. In addition, it has been shown that these compounds also facilitate root colonization (Stein 2005) and reinforce the host’s resistance potential (Ongena et al. 2007). In Musa plants inoculated with M. fijiensis and in the presence of CF, typical symptoms of disease were observed as previously described by Alvarado-Capó et al. (2003). The incubation time for all treatments was 14 days. In all cases, the pathogen completed the cycle of disease and necrotic lesions with dry center were observed in the evaluation period (Fig. 3a). However, the inoculated plants with the CF applied Fig. 3 Effect of culture filtrate of Bacillus pumilus CCIBP-C5 in artificially inoculated banana plants with Mycosphaerella fijiensis under greenhouse conditions. a) Banana plants artificial inoculated with M. fijiensis at 70dpi, b) Area under disease progress curve (AUDPC) (from 21 to 70 dpi). Different letters on bars indicate significant differences by Kruskal-Wallis / Mann-Whitney, p <0.05 (n = 10). c) Number of necrotic lesions at 49 dpi. Different letters on bars indicate significant differences by Kruskal-Wallis / Mann-Whitney, p <0.05 (n = 10). (Control) plants inoculated with M. fijiensis suspension, (CF B) CF applied in the abaxial surface of the leaf at 72 h before M. fijiensis inoculation, (CF A) CF applied at 72 h after M. fijiensis inoculation Author's personal copy Trop. plant pathol. (2017) 42:121–125 three days after inoculation did not reach the state 5 of symptoms according to the qualitative scale used. On the other hand, plants inoculated with M. fijiensis (control) completed the disease cycle 42 days after inoculation. In comparison, treatments with CCIBP-C5 CF showed a significant disease reduction. The CF B treatment delayed the disease progress by 21 days, while treatment CF A did it by 28 days with respect to control. Comparisons of AUDPC values indicated that the CF applied after inoculation of M. fijiensis showed significant differences with respect to the control (Fig. 3b). This variable quantifies the amount of disease in a given period and its lower value with respect to the control indicates a pathogen reduction on the plant or a reduction of the initial inoculum. Also, fewer necrotic lesions were observed in treatment with CF after pathogen inoculation (86% lower compared to control) (Fig. 3c). Moreover, with a single application of CCIBP-C5 CF (after pathogen inoculation) the efficiency of disease control was of 33.6% at 70 dpi. The CF did not contain preservatives or stabilizers that protect it from the action of sunlight, water irrigation or other environmental factors it was exposed to, which indicates its potential. Until now, similar results related to biological control of bacterial culture filtrate against BLSD have not been reported. The results of this study expand our understanding of the mechanisms by which B. pumilus CCIBP-C5 exerts antifungal activity against M. fijiensis. The biocontrol effects of CFCCIBP-C5 on banana plants artificially inoculated with M. fijiensis might be attributed to production of antifungal metabolites with negative effects on hyphal tip structure and membrane damage. The characteristics of this strain and its metabolites, make it a promising agent for the biological control of BLSD on banana. Acknowledgements The authors are grateful to MSc. Miguel Tzec for his help in microscopy technics. Partial financial support for this work was an output of a scholarship from the Food Security Center from the University of Hohenheim, which is part of the DAAD (German Academic Exchange Service) program “exceed” in cooperation with the host institute Centro de Investigación Científica de Yucatán (CICY), México and PhD Blondy Canto Canché. References Alvarado-Capó Y, Leiva-Mora M, Rodríguez MA, Acosta-Suárez M, Cruz-Martín M, Portal O, Kosky RG, García L, Bermúdez I, Padrón J (2003) Early evaluation of black leaf streak resistance by using mycelial suspensión of Mycosphaerella fijiensis. In: Jacome L, Lepoivre P, Martin D, Ortiz R, Romero R, Escalante JV (eds) Mycosphaerella leaf spot diseases of bananas: present status and outlook. Proceedings of the Workshop on Mycosphaerella Leaf Spot Diseases, San José, Costa Rica. The International Network for the Improvement of Banana and Plantain, Montpellier, pp 169– 175 125 Ceballos I, Mosquera S, Angulo M, Mira J, Argel L, Uribe-Velez D, Romero-Tabarez M, Orduz-Peralta S, Villegas V (2012) Cultivable bacteria populations associated with leaves of banana and plantain plants and their antagonistic activity against Mycosphaerella fijiensis. Microb Ecol 64:641–653 Choi G, Lee HJ, Cho K (1996) Lipid peroxidation and membrane disruption by vinclozolin in dicarboximide-susceptible and resistant isolates of Botrytis cinerea. Pest Biochem Physiol 55:29–39 Churchill AC (2011) Mycosphaerella fijiensis, the black leaf streak pathogen of banana: progress towards understanding pathogen biology and detection, disease development, and the challenges of control. Mol Plant Pathol 12:307–328 Cruz-Martín M, Acosta-Suárez M, Mena E, Roque B, Leiva-Mora M, Pichardo T, del Pilar-Castro R, Alvarado-Capó Y (2013) Cuantificación del crecimiento in vitro de Mycosphaerella fijiensis mediante lecturas de absorbancia. Biotecnología Vegetal 13:219– 224 Fu G, Huang S, Ye Y, Wu Y, Cen Z, Lin S (2010) Characterization of a bacterial biocontrol strain B106 and its efficacy in controlling banana leaf spot and post-harvest anthracnose diseases. Biol Control 55:1–10 Gutierrez-Monsalve JA, Mosquera S, González-Jaramillo LM, Mira JJ, Villegas-Escobar V (2015) Effective control of black Sigatoka disease using a microbial fungicide based on Bacillus subtilis EACB0015 culture. Biol Control 87:39–46 Gutiérrez-Román MI, Holguín-Meléndez F, Dunn F, Guillén K, HuertaPalacios G (2015) Antifungal activity of Serratia marcescens CFFSUR-B2 purified chitinolytic enzymes and prodigiosin against Mycosphaerella fijiensis, causal agent of black Sigatoka in banana (Musa spp.). BioControl 60:565–572 Johanson LH, Jeger MJ (1993) Use of PCR for detection of Mycosphaerella fijiensis and musicola, the causal agents of Sigatoka leaf spots in banana and plantain. Mycol Res 97:670–674 Leiva-Mora M, Alvarado-Capó Y, Acosta-Suárez M, Cruz-Martín M, Sánchez-García C, Roque B (2010) Protocolo para la inoculación artificial de plantas de Musa spp. con Mycosphaerella fijiensis y evaluación de su respuesta mediante variables epifitiológicas y componentes de la resistencia. Biotecnología Vegetal 10:79–88 Matteuci E, Biasci E, Giampietro O (2001) Advanced oxidation protein products in plasma: stability during storage and corrrelation with other clinical characteristics. Acta Diabetol 38:187–189 Moreno B, Dunn F, Guillén K, Holguín F, Hernández M, Encarnación S, Huerta G (2016) Antifungal performance of extracellular chitinases and culture supernatants of Streptomyces galilaeus CFFSUR-B12 against Mycosphaerella fijiensis Morelet. World J Microbiol Biotechnol 32:44 Ongena M, Jourdan E, Adam A, Paquot M, Brans A, Joris B (2007) Surfactin and fengycin lipopeptidesof Bacillus subtilis as elicitors of induced systemic resistance in plants. Environ Microbiol 9:1084– 1090 Ruangwong O, Chang C, Lamine S, Liang W (2012) Identification of antifungal compound produced by Bacillus subtilis LB5 with ability to control anthracnose disease caused by Colletotrichum gloeosporioides. Afr J Microbiol Res 6:3732–3738 Shanner G, Finney RE (1977) The effect of nitrogen fertilization on the expression of show-mildewing resistance in Knox wheat. Phytopathology 67:1051–1056 Stein T (2005) Bacillus subtilis antibiotics: structures, syntheses and specific functions. Mol Microbiol 56:845–857 Wang C, Xing J, Chin CK, Peters JS (2002) Fatty acids with certain structural characteristics are potent inhibitors of germination and inducers of cell death of powdery mildew. Physiol Mol Plant Pathol 61:151–161 Zhou W, Huang J, Niu T (2008) Isolation of an antifungal Paenibacillus strain HT16 from locusts and purification of its medium dependent antagonistic component. J Appl Microbiol 105:912–919