Characterization of a bacterial biocontrol strain B106 and its efficacies on controlling banana leaf spot and post-harvest anthracnose diseases

Biological Control 55 (2010) 1–10

Contents lists available at ScienceDirect

Biological Control journal homepage: www.elsevier.com/locate/ybcon

Characterization of a bacterial biocontrol strain B106 and its efficacy in controlling banana leaf spot and post-harvest anthracnose diseases Gang Fu a,b, Siliang Huang c,*, Yunfeng Ye a, Yongguan Wu a, Zhenlu Cen b, Shanhai Lin a a

College of Agriculture, Guangxi University, Nanning 530005, PR China Microbiology Research Institute, Guangxi Academy of Agricultural Sciences, Nanning 530007, PR China c College of Life Sciences and Technology, Nanyang Normal University, Nanyang 473061, PR China b

a r t i c l e

i n f o

Article history: Received 5 August 2009 Accepted 4 May 2010 Available online 9 May 2010 Keywords: Bacillus subtilis Biological character Musa Banana leaf spot Pseudocercospora musae Banana anthracnose Colletotrichum musae

a b s t r a c t An antagonistic bacterial strain B106 was isolated from the rhizospheric soil of a banana plant in Nanning city, Guangxi, China, and identified as Bacillus subtilis based on its 16S rDNA sequence homology with the related bacteria from GenBank as well as physiological and biochemical characters. The cultural conditions were optimized for enhancing the efficacy of the antagonist against banana leaf spot caused by Pseudocercospora musae (teleomorph: Mycosphaerella musicola) and post-harvest anthracnose by Colletotrichum musae. The optimized cultural condition for strain B106 to express higher antagonistic activity against P. musae was the combination of 31 °C, pH 6.0, EM medium and 5-day-incubation. However, the optimized cultural condition for the bacterium to produce higher biomass was the combination of 31–34 °C, pH 6.5, EM medium and 3-day-incubation. The results based on greenhouse tests showed that 72.3% efficacy of the antagonist in controlling the banana leaf spot disease was obtained 10 days after pathogen inoculation. The efficacy of strain B106 (1  108 CFU ml1) in controlling both banana leaf spot diseases in the field and anthracnose disease at post-harvest stage was 48.3% and 48.6%, respectively, under the optimized cultural condition for the strain to express higher antagonistic activity. The experimental data indicated that the antagonistic strain was a promising biocontrol agent against the banana diseases. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction Banana is the most popular fruit crop commercially grown in many tropical and subtropical countries for its utilization as dessert and as staple food. It is one of the most important fruit crops in international trade for earning foreign exchange in many African countries. The infectious diseases caused by pathogenic microbes, such as fungi, bacteria and viruses, are the major limiting factor in successful quality production of this crop. Among these pathogen groups, fungi are the most important members involved in pre- and post-harvest banana infectious diseases. The fungus-induced banana leaf spots are the economically important diseases in banana production worldwide. The most serious banana leaf spots are caused by three species of Mycosphaerella: Mycosphaerella musicola, Mycosphaerella fijiensis and Mycosphaerella eumusae (Jones, 2003). It has been estimated that the black Sigatoka (black leaf streak) caused by M. fijiensis could result in 33–69% yield loss on plantain and banana (Bureau, 1990; Romero, 1986; Mobambo et al., 1993, 1996; Marín et al., 2003), which accounts for at least 27% of the total cost of disease control (Stover, 1980). Yellow * Corresponding author. Fax: +86 377 63513461. E-mail address: [email protected] (S. Huang). 1049-9644/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.biocontrol.2010.05.001

Sigatoka disease caused by Pseudocercospora musae (teleomorph: M. musicola), was first reported in Java in 1902 (Zimmerman, 1902), and has broken out in various locations in Asia, Africa, and Latin America from 1930 s (Hayden et al., 2003). Besides the three Mycosphaerella spp. described above, there are many other fungal species causing banana or plantain leaf spots, such as Deightoniella torulosa (Koné et al., 2008), Bipolaris sacchari (Silva et al., 2008), Pestalotiopsis menezesiana (Huang et al., 2007), Cladosporium musae (Surridge et al., 2003), Cordana musae, Helminthosporium torulosa, Curvularia lunata, and Alternaria musae (Jones, 2000). In China, at least 12 pathogenic fungi are known to be the causal agents of banana leaf spots in the field. At present, yellow Sigatoka disease has become predominated among the banana leaf spot diseases in South China (unpublished data) since first recorded in 1936 (Meredith, 1970). On the other hand, although Colletotrichum musae occasionally causes minor banana leaf spots in the field in South China (unpublished data), its economically significant damage to banana crop mainly occurs at post-harvest stage. C. musae forms quiescent infections on unmatured banana fruits in the field and causes brown fruit rot (anthracnose) during ripening (de Lapeyre de Bellaire et al., 2000), which leads to a loss of 10–86% of the banana fruits during shipment and storage (Krauss and Johanson, 2000; Alvindia et al., 2000).

2

G. Fu et al. / Biological Control 55 (2010) 1–10

Fungicides are the primary means of controlling banana leaf spots and post-harvest anthracnose diseases. Several types of fungicides have been used for suppressing the banana diseases, such as triazoles (Romero and Sutton, 1997), strobilurins (Chin et al., 2001; Pérez et al., 2002), spiroketalamins (Khan et al., 2001) and imidazoles (Koné et al., 2009; Washington, 1997; Washington et al., 1998; Krauss and Johanson, 2000). At present, fungicide resistance has been detected for benomyl (Romero and Sutton, 1998), thiabendazole (de Lapeyre de Bellaire and Dubois, 1997), azoxystrobin (Amil et al., 2007), trifloxystrobin (Pérez et al., 2002), propiconazole (Romero and Sutton, 1997) and trifloxystrobine (Chin et al., 2001) due to the extensive and frequent usage of the chemicals for the control of pre- and/or post-harvest banana diseases. Although the recently developed fungicides such as triazoles and prochloraz are highly effective for suppressing these diseases, the overuse of fungicides might result in chemical residues in banana plants and their environments which become a public concern. For sustainable development of banana production, it is necessary to develop ecofriendly methods for reducing fungicide usage in the management of banana diseases. For many years, breeding for Sigatoka resistance has been considered as an important strategy to control the disease. Harelimana et al. (1997) used the toxins produced by M. fijiensis for the screening of banana cultivars resistant to yellow Sigatoka of bananas. Twizeyimana et al. (2007) reported two rapid assays, using plantlets in tubes and detached leaves, for the screening of Musae species resistant to black Sigatoka. Furthermore, a few natural products have been used for the control of the banana diseases. Vawdrey et al. (2004) evaluated the effectiveness of five mineral oils, four plant oils and a plant-derived sticker on suppressing yellow Sigatoka of bananas. In the case of banana post-harvest anthracnose, several essential oils were developed to control the disease (Ranasinghe et al., 2002, 2005; Demerutis et al., 2008; Anthony et al., 2004). In addition, hot water treatment has been regarded as another effective way for controlling this disease (Reyes et al., 1998; Win et al., 2007). The application of biocontrol agents that are safe for the environment has become an important strategy in integrated pest management for enhancing plant productivity and safety of agro-products. Several antagonists such as Trichoderma harzianum (Alvindia and Natsuaki, 2008), Burkholderia cepacia (De Costa and Erabadupitiya, 2005), Bacillus amyloliquefaciens (Alvindia and Natsuaki, 2009), Pseudomonas syringae (Williamson et al., 2008), Pichia anomala and Candida oleophila (Lassois et al., 2008), Pantoea agglomerans and Flavobacterium sp. (Niroshini Gunasinghe and Karunaratne, 2009) were selected for controlling banana post-harvest diseases. A few biocontrol agents have been reported for the suppression of black Sigatoka of bananas in greenhouse conditions, but have not been applied in the field yet (Jiménez et al., 1987; González et al., 1996; Miranda Corrales, 1996). An antagonistic bacterial strain B106, isolated from the rhizospheric soil of a banana plant in Nanning city, Guangxi, China, showed strong antagonistic activity against multiple pathogenic fungi causing banana pre- and post-harvest diseases in vitro (Fu et al., 2007). The inhibition rates of the strain have been found to be approximately 89% against C. musae, and more than 70% against four banana leaf spot pathogens P. ( musae, C. musae, H. torulosa and C. lunata). In addition, an inhibition rate of greater than 50% has also been found to suppress another banana leaf spot pathogen (A. musae) and one of the banana crown rot pathogens (Fusarium roseum) (Fu et al., 2007). Our previous in vitro investigations on the suppressive activities of strain B106 against the major pathogens causing banana leaf spots and post-harvest anthracnose suggested that the strain might be a promising biocontrol agent against these diseases. However, the classification position and biological characters of the antagonistic strain are unclear. The optimization of cultural conditions for the antagonistic strain to

express high antagonistic activity and biomass production are needed for enhancing its efficacy on field utilization. The objectives of the study are: (1) to characterize and identify strain B106 using bacteriological and molecular methods; (2) to optimize cultural conditions of strain B106 for an enhanced control efficacy (CE); and (3) to evaluate the efficacy of strain B106 in controlling both banana leaf spot diseases in the field and banana anthracnose at the post-harvest stage. 2. Materials and methods 2.1. Plant materials, pathogens and biocontrol strains 2.1.1. Plant materials and fruits used Both the banana plants and fruits used were Williams (AAA, Cavendish type), a popular banana cultivar in South China. The cultivar was susceptible to banana leaf spot and anthracnose diseases. The tissue culture plantlets (20-day-old) were obtained from Biotechnology Research Institute, Guangxi Academy of Agricultural Sciences, Nanning, China and grown in nutrition pots in a greenhouse until used. Entire bunches of fungicide-free banana fruits with approximately 80% maturity were purchased from the banana growers in Jinling township, Nanning city, and used for post-harvest experiments. 2.1.2. Preparation of pathogen Five pathogens causing banana leaf spot diseases (P. musae, C. musae, H. torulosa, C. lunata and A. musae) were isolated from affected banana leaves in Guangxi, China, in 2005. C. musae and F. roseum were isolated from the diseased banana fruits in 2006. Affected banana leaf or fruit samples (cultivar Williams) were collected from the banana plantations at Jinling township in Nanning city. The samples were washed under running tap water. Cut tissues (0.5  0.5 cm2) were surface-disinfested with 0.1% sodium hypochloride for 2–3 min, rinsed with sterilized water twice and placed on potato–dextrose agar (PDA: 200-g-potato-boiled extract, 200 ml; dextrose, 20 g; agar, 15 g; distilled water, 800 ml; pH 7.0) plates. After 4 days of incubation at 28 °C, growing mycelia from the sample tissues were removed and purified by single spore isolation (Huang and Kohmoto, 1991). The pathogenicity of the isolated fungi to banana leaves or fruits were confirmed by inoculation tests. All of the banana pathogens used in the study were stored in 25% glycerol at 20 °C. Working cultures were stored on PDA slants at 4 °C. Conidial suspension of P. musae was prepared by flooding the 14-day-old colonies with sterilized distilled water and agitated with a glass rod. The suspension was filtered through a sterilized muslin cloth. Conidia were harvested in an Erlenmeyer flask, and counted with a haemacytometer (Hausser, Horsham, PA, USA). The spore concentration was adjusted to 105 ml1. P. musae was a predominant member of the causal agents of leaf spots in the main banana-producing areas in South China, and used as the target pathogen for greenhouse trial. 2.1.3. Isolation of bacterial biocontrol agents Bacterial biocontrol agents were isolated from the rhizospheric soil of banana plants in Nanning city. The soil sample (approximately 10 g) was transferred into a 50-ml-Erlenmeyer flask containing 10 ml sterilized water and shaken at 200 rpm for 20 min. The soil suspension was then diluted to 102, 103, 104, and 105 times, respectively. A loopful of the soil suspension was streaked on beef–peptone–yeast–dextrose agar (BPYDA: beef extract, 3 g; peptone, 5 g; yeast extract, 1 g; dextrose, 10 g; agar, 15 g; distilled water, 1000 ml; pH 7.0) plates. After being incubated at 28 °C for 24 h, a single bacterial colony well separating from the other ones was restreaked on a new BPYDA plate with an inoculating loop.

G. Fu et al. / Biological Control 55 (2010) 1–10

The final single colony was then transferred on BPYDA slants and stored at 4 °C until used. 2.1.4. Screening and selection of antagonistic agents Bacterial isolates were evaluated in vitro for their ability to suppress the growth of P. musae, using the dual incubation method (Fokkema, 1978). Bacteria resulting in more than 30% growth inhibition (GI) of P. musae were further tested for antibiosis toC. musae, H. torulosa, C. lunata, A. musae, C. musae and F. roseum. A loopful of bacteria grown in BPYDA for 24 h was stab-inoculated at three equidistant points 1 cm from the plate periphery (9 cm in diameter), while a mycelial plug of the test pathogens was placed in the center of PDA plates. Mycelial plugs, 6 mm in diameter, were taken from the edge of 7-day-old fungal colonies maintained on PDA plates. Plates inoculated only with the pure pathogens served as control(CK). All of the plates were incubated at 28 °C. Each treatment consisted of three replicates. Percent GI was determined after 10 days of incubation using the following formula (Korsten et al., 1995): Kr  r1/Kr  100 = GI, where Kr represents the distance from the point of inoculation to the colony margin on CK plates, r1 the distance of fungal growth from the point of inoculation to the colony margin on treated plates in the direction of the antagonist, and GI the percent growth inhibition. The bacteria with more than 30% GI against the seven tested pathogens were selected as candidate biocontrol agents for further studies. Strain B106 was then selected for further laboratory and field studies for its stable and higher inhibition ability against the test pathogens. 2.1.5. Preparation of antagonistic bacterium The strain B106 was maintained on BPYDA at 28 °C and stored at 4 °C. When a liquid culture was needed, strain B106 was grown in BPYDB (BPYDA except agar) or Emerson medium (EM: dextrose, 10 g; beef extract, 4 g; NaCl, 2.5 g; yeast extract, 10 g; peptone, 4 g; distilled water, 1000 ml; pH 7.0) under a desired cultural condition. The OD (optical density) values of the resultant cell suspensions at 625 nm were determined using an UV/Vis spectrophotometer (UV-1600, Beijing Rayleigh, China). The bacterial concentration was adjusted to approximately 108–109 colony forming unit (CFU) ml1 with sterilized water. 2.2. Identification of strain B106 Strain B106 was preliminarily characterized by means of conventional bacteriological methods including morphological, physiological and biochemical characters (Buchanan and Gibbons, 1984). Strain B106 was grown on BPYDA plates for 24 h. The morphological characteristics of the bacterium was observed with a transmission electron microscope (HITACHI, H-500) as described by Williams et al. (1990). Colony morphological descriptions were done after incubation on BPYDA plates at 28 °C for 24–48 h. Other bacteriological tests including Gram-staining, bacterial spore formation, catalase reaction, Voges–Proskauer reaction, salt tolerance, acid production (from D-glucose, L-arabinose, D-xylose, and D-mannitol), glutin hydrolysis, starch hydrolysis, utilization of sodium citrate, nitrate reduction, milk peptonization, sulfureted hydrogen, methyl red, indole, growth without oxygen, gas production from D-glucose, decomposition of tyrosine, egg yolk agar and growth at pH 5.7, pH 6.8, 50 °C were carried out based on the general bacterial experiment methods (Buchanan and Gibbons, 1984; Williams et al., 1990; Ayyadurai et al., 2006). Strain B106 was further identified based on comparative sequence analysis of its 16S rDNA gene as follows. The strain was incubated in BPYDB at 28 °C for 16 h prior to the extraction of its genomic DNA. The total genomic DNA was extracted using the DNA Extraction Kit (Bioer Technology Co., Ltd., China). The PCR reaction (50 ll) mixture contained 5 ll of 10  PCR

3

buffer, 1 ll of dNTPs (25 mM), 2 ll of the forward primer F27 (50 -AGAGTTTGATCATGGCTCAG-30 ), 2 ll of the reverse primer R1492 (50 -TACGGTTACCTTGTTACGACTT -30 ), 0.4 ll of Taq DNA polymerase (5 U ll1), 4 ll of MgCl2 (25 mM), 1 ll of DNA template and 34.6 ll of autoclaved Milli-Q water. The PCR was performed on an automated thermocycler device (TC-16/H, Hangzhou ThermoMagnetics Co., Ltd.) using the following parameters: denaturation at 94 °C for 4 min followed by 35 cycles of denaturation at 94 °C for 40 s, annealing at 50 °C for 50 s, elongating at 72 °C for 80 s and a final extension at 72 °C for 8 min. The PCR products were detected by gel electrophoresis and purified with the PCR Product Purification Kit (Takara Biotechnology Co., Ltd.). The purified products were sent to Shanghai Sangon Biological Engineering Technology & Services Co., Ltd. for sequencing (both strands sequenced). The 16S rDNA sequence of strain B106 was analyzed and aligned with the related sequences retrieved from GenBank database using NCBI BLAST searching tool. A phylogenetic tree based on 16S rDNA sequences was constructed with the neighbor-joining algorithm in MEGA version 2.1. 2.3. In vitro inhibitory ability of strain B106 against P. musae Strain B106 was grown in BPYDB on a rotary shaker (150 rpm) at 28 °C for 3 days and the resultant cell suspension adjusted to a concentration of 1  109 CFU ml1. The inhibitory activities of the cell suspensions against P. musae were tested using the dual incubation method (Fokkema, 1978). The antagonistic test was repeated three times under the same conditions. The GI was investigated as the colony of P. musae extended to the margin of the CK plate (90 mm diameter). 2.4. Effects of cultural conditions on biomass and inhibitory activities of strain B106 against P. musae 2.4.1. Incubation time Strain B106 was incubated in BPYDB on a rotary shaker (150 rpm) at 28 °C for eight different time courses ranged from 24 to 192 h. The OD625 values of the resultant cell suspensions were separately measured at each incubation time using the spectrophotometer. The inhibitory activities of the cell suspensions against P. musae at each incubation time were separately examined using the dual incubation method (Fokkema, 1978). 2.4.2. Incubation temperature The cell suspension of strain B106 (40 ll, 1  109 CFU ml1) was inoculated in 100 ml BPYDB and incubated at temperatures ranged from 10 to 43 °C with 3 °C intervals. The cell suspension (5 of 100 ml) was taken out in a superclean bench and its OD value at 625 nm measured 24 h after inoculation. The remaining 95 ml of the cell suspension was subsequently incubated for another 48h-period prior to the examination of its inhibitory activity against P. musae. The examination of the inhibitory activity was conducted using the dual incubation method (Fokkema, 1978). 2.4.3. Medium pH Strain B106 was inoculated in BPYDB with initial pHs ranged from 3.0 to 11.0 (0.5 interval) and incubated under agitation (150 rpm) at 28 °C. The OD values of 24-h-old cell suspensions were measured with the spectrophotometer. The inhibitory activities of 72-h-old cell suspensions against P. musae were examined using the dual incubation method (Fokkema, 1978). 2.4.4. Nutrition Strain B106 was separately inoculated in the following liquid media: potato–dextrose broth (PDB: PDA except agar), BPYDB, Gause’s No. 1 (KNO3, 1 g; K2HPO4, 0.5 g; MgSO4, 0.5 g; NaCl, 0.5 g;

4

G. Fu et al. / Biological Control 55 (2010) 1–10

FeSO4, 0.01 g; soluble starch, 20 g; distilled water, 1000 ml), Czapek (sucrose, 30 g; NaNO3, 2 g; K2HPO4, 1 g; MgSO4, 0.5 g; KCl, 0.5 g; FeSO4, 0.01 g; distilled water, 1000 ml), Keshi No. 1 (K2HPO4, 1 g; MgCO3, 0.3 g; NaCl, 0.2 g; KNO3, 1 g; FeSO4, 0.01 g; CaCO3, 0.5 g; dextrose, 20 g; distilled water, 1000 ml), EM, potato broth (PB: extract of 200-g-boiled potato, 200 ml; distilled water, 800 ml), dextrose solution (DS: dextrose, 20 g; distilled water, 1000 ml), starch solution (SS: soluble starch, 30 g; distilled water, 1000 ml), banana–dextrose broth (BDB: 50-g-banana-leaf-boiled extract, 200 ml; dextrose, 20 g; distilled water, 1000 ml), asparagine–dextrose solution (ADS: aspartoyl, 0.5 g; dextrose, 10 g; K2HPO4, 0.5 g; distilled water, 1000 ml), yeast–dextrose broth (YDB: yeast extract, 10 g; dextrose, 20 g; distilled water, 1000 ml), soybean broth (SB: soybean flour, 30 g; distilled water, 1000 ml), oat broth (OB: 30-g-oat-boiled extract, 200 ml, distilled water, 800 ml), corn broth (CB: 30-g-cornmeal-boiled extract, 200 ml; distilled water, 1000 ml) and starch– salt solution [SSS: soluble starch, 10 g; (NH4)2SO4, 2 g; K2HPO4, 1 g; MgSO4 1 g, NaCl, 1 g; CaCO3, 3 g; distilled water, 1000 ml] under agitation (150 rpm) at 28 °C. The OD values of 24-h-old cell suspensions were measured with the spectrophotometer. The inhibitory activities of 72-h-old cell suspensions against P. musae were determined using the dual incubation method (Fokkema, 1978). 2.5. Efficacy of strain B106 in controlling yellow Sigatoka in greenhouse The 7-week-old banana tissue culture plantlets with approximately 30 cm plant height were transplanted into 26-cm-diameter pots (one plant per pot) in greenhouse for 14 days. P. musae was used as the pathogen for the greenhouse test. The temperatures were 26– 32 °C and relative humidities (RH) 76–82% in the greenhouse during the experiment period (from August 3 to 18 of 2007). Each treatment consisted of 18 plants with 3 replicates (6 plants per replicate). The experiment was performed as follows. Three fully-expended top leaves per plant were slightly wounded by stinging equably (4 point per cm2) with sterilized needles before pathogen inoculation. The conidial suspension (1  105 spores ml1) of P. musae was sprayed on both sides of the leaves (2 ml per leaf), using an artist’s airbrush (Apoho 16B, Suzhou Plastic Electron Co., Ltd., China). The strain B106 was grown in EM under agitation (150 rpm) at 31 °C for 3 days and its cell suspension adjusted to a concentration of 1  108 CFU ml1. Strain B106 and propiconazole were separately sprayed on the leaves using the same method as in the treatment of P. musae. To evaluate the effect of spray timing on control efficacy of strain B106 against the disease, the treatments were set up as follows: SB, strain B106 applied on the banana leaves 2 days before P. musae inoculation; SA, strain B106 applied on the banana leaves 2 days after P. musae inoculation; PA, propiconazole (375 mg kg1), a highly effective fungicide commonly used for controlling the banana leaf spots in the experimental area, was applied 2 days after P. musae inoculation. The banana plants inoculated with the conidial suspension of P. musae were used as CK. The disease severity (DS), disease index (DI) and greenhouse control efficacy (GCE) were separately investigated 10 and 14 days after P. musae inoculation. The DS, expressed as a percentage of infected area over the total leaf area, was divided into five ratings: 0 = no symptom, 1 = less than 5% leaf area infected, 3 = 5–15% leaf area infected, 5 = 16– 25% leaf area infected, 7 = 26–50% leaf area infected, 9 = more than 50% leaf area infected. The DI and GCE were calculated using following formulae:

h i X DI ¼ 100  ðNo: of affected leaves  corresponding DSÞ

=ðNo: of total leaves  9Þ GCE ð%Þ ¼ 100  ðDI of CK  DI of B106 or fungicide treatmentÞ =DI of CK

2.6. Field trials 2.6.1. Efficacy of strain B106 in controlling banana leaf spots in the field To evaluate CE of strain B106 against banana leaf spots, two trials were carried out in the same field (120 m above MSL, soil pH: 6.7, soil type: sandy clay loam, clay: 18.2%, silt: 32.4%, sand: 42.3%, organic matter: 5.8%) cultivated with tissue culture banana plants cv. Williams (AAA) in Jinling township, Nanning city, Guangxi, China. The first field trial was carried out from June 23 to July 31 in 2007. Banana plants were grown during the trial under the climatic conditions: rain fall 489 mm; mean temperature 21.4 °C and RH 78–91%. The experimental design was a complete randomized block with a 1.8-m-planting spacing. Each treatment consisted of 20 plants with 4 replicates (5 plants per replicate). Strain B106 was grown in BPYDB under agitation (150 rpm) at 28 °C for 2 days. Two treatments (strain B106 and propiconazole) were set up in the trial. The cell suspension (1  108 CFU ml1) of strain B106 and propiconazole (375 mg kg1) were separately sprayed on the banana leaves. Clean water was sprayed on the CK blocks. All treatments were carried out with a knapsack mist blower sprayer (ARDLEX, HD-400, Brazil) in afternoon. Both sides of the banana leaves were sprayed to the point of runoff (500–700 ml liquid per plant). The first application was performed at 28th week after transplanting when the leaf spot diseases first appeared. The application was performed three times at 14-day-intervals. The DS on the 13 fully-expanded leaves from the top of a banana plant was investigated and DI calculated. The DS investigation was performed at each application time and the final investigation performed 10 days after the last application. The DI of the CK determined at the first application was used as the base number (BN) for evaluating field control efficacy (FCE). FCE was calculated using following formula:

FCE ð%Þ ¼ ½1  ðBN  DI of treatment block at the 2nd or 3rd applicationÞ=ðDI of treatment block at the 1st application  DI of CK block at the 2nd or 3rd applicationÞ  100% The second field trial was conducted from June 28 to August 5 in 2008 at the location where the first trial performed. Banana plants were grown during the trial under the climatic conditions: rain fall 445 mm, mean temperature 20.7 °C and RH 76–90%. The experimental design of the second trial was the same as the first one except for the cultural conditions of strain B106. In this trial, strain B106 was grown under agitation (150 rpm) and the optimized cultural conditions including the optimal growing medium (EM), initial pH value (pH 6.5), temperature (31 °C) and incubation time (3 days). 2.6.2. Efficacy of strain B106 in controlling banana post-harvest anthracnose The cell suspension of strain B106 was prepared under the optimized cultural conditions described above. Freshly harvested nonsymptomatic banana fruits were gently washed with tap water and air-dried. Two treatments were set up and each treatment consisted of four replicates (5 ± 0.1 kg banana fruits per replicate). The banana fruits were separately dipped in a cell suspension (1  108 CFU ml1) of strain B106 for 10 min and in prochloraz (375 mg kg1) for 1 min. The fruits dipped in tap water were used as CK. After being treated with the antagonist, the fruits of each replicate were packed in a polyethylene bag and maintained at 28 ± 1 °C and RH 78 ± 2% until symptoms developed. The DS, DI and CE were separately investigated 14 and 17 days after the antagonist application.

G. Fu et al. / Biological Control 55 (2010) 1–10

The DS, expressed as a percentage of lesion area over the total surface area per fruit, was divided into five ratings: 0 = no symptom, 1 = less than 10% fruit surface area infected, 3 = 11–25% fruit surface area infected, 5 = 26–50% fruit surface area infected, 7 = 51–75% fruit surface area infected, 9 = more than 75% fruit surface area infected. The DI and CE were calculated using following formulae:

h i X DI ¼ 100  ðNo: of diseased fruits  corresponding DSÞ

2.7. Statistical analysis Data were subjected to analysis of variance (ANOVA) using SAS software (SAS Institute, version 6.08, Cary, NC). Statistical significance was assessed using Duncan’s Multiple Range Test at P = 0.05. 3. Results 3.1. Identification of strain B106

=ðNo: of total fruits  9Þ CE ð%Þ ¼ 100  ðDI of CK  DI of B106 or fungicide treatmentÞ =DI of CK Table 1 Physiological and biochemical characters of strain B106.

a

5

Item tested

Reactiona

Gram stain Catalase Voges–Proskauer Acid production from D-glucose Acid production from L-arabinose Acid production from D-xylose Acid production from D-mannitol Glutin hydrolysis Hydrolyzation of amylum Utilization of sodium citrate Nitrate reduction Milk peptonization Growth at pH 6.8 Growth at pH 5.7 Growth at 50 °C Growth in 7% NaCl Sulfureted hydrogen Methyl red Indole Growth without oxygen Gas production from D-glucose Decomposition of tyrosine Egg yolk agar

+ + + + + + + + + + + + + + + + + +     

‘‘+” and ‘‘” represent positive and negative reactions, respectively.

Strain B106 grew on BPYDA with a subround and straw yellow colony. It was a spore-forming rod bacterium with peritrichous flagella. The dimension of the bacterium was 0.6–0.9 lm  2.1– 3.5 lm. As shown in Table 1, the physiological and biochemical characters of the antagonist were in complete correspondence with Bacillus subtilis (Buchanan and Gibbons, 1984). The 16S rDNA of strain B106 was amplified with primers F27 and R1492, and its sequence submitted to GenBank database (accession number: FJ598009). The analysis of 16S rDNA sequence indicated that the strain shared 99% maximal identity with B. subtilis. Moreover, strain B106 clustered with two B. subtilis strains from GenBank in the phylogenetic tree (Fig. 1), clearly demonstrating that the strain was a member of B. subtilis at 16S rDNA sequence homology level. 3.2. In vitro inhibitory activity of strain B106 against P. musae By co-incubating strain B106 with P. musae on the same PDA plate at 28 °C for 7 days, the former strongly inhibited growth of the latter. The average inhibition rate of the antagonist against P. musae was 84.7%. 3.3. Effects of cultural conditions on biomass and inhibitory activity of strain B106 against P. musae The bacterial biomass expressed by OD values at 625 nm (OD625) and the inhibition rates of strain B106 against P. musae obviously increased after 3 days of incubation and subsequently

Fig. 1. Neighbor-joining phylogenetic tree based on 16S rDNA sequences showing the relationships between strain B106 and related bacteria from GenBank database. The numbers in parentheses represent the sequence accession numbers in GenBank. The values of 50 or more (from 1000 replicates) are indicated at the branch nodes as the percentages supported by bootstrap. The sequence of Agrobacterium rhizogenes (FJ527681) was used as an outgroup. The scale bar represents 0.02 substitutions per nucleotide position.

6

G. Fu et al. / Biological Control 55 (2010) 1–10

Fig. 2. Effects of cultural conditions (A, incubation time; B, incubation temperature; C, medium pH; D, nutrition) on biomass (OD625) and inhibitory activity (inhibition rate) of strain B106 against P. musae. In the experiment of A, the OD625 values and inhibition rates were separately measured at each incubation time (24-h-interval). In the experiments of B, C and D, the OD625 values and inhibition rates were measured 24 and 72 h after incubation, respectively.

maintained at a stable level. The highest biomass(OD625 = 2.18) and peak inhibition rate (85.7%) occurred at the 3rd and 5th days of incubation, respectively (Fig. 2A). Strain B106 could grow at temperatures ranged from 13 to 43 °C and all of the resultant cell suspensions showed inhibitory activities against P. musae at different levels. The highest biomass (OD625 = 2.24) and peak inhibition rate (73.4%) were observed at 31 °C. The temperature range for the strain to express higher biomass and stronger inhibitory activities was 31–34 °C. Within the temperature range of 10–31 °C, both the biomass and inhibition rate raised with the increase of temperature. As the temperature exceeded 31 °C, both the biomass and inhibition rate decreased. Neither the growth of the antagonist nor the inhibitory activity of the inoculated medium against P. musae was found at 10 °C (Fig. 2B). The pHs for strain B106 to grow and express inhibitory activities were 3.0–11.0. The highest biomass (OD625 = 2.04) and peak inhibition rate (85.1%) were observed at pH 6.5 and pH 6.0, respectively. pH 5.0–8.0 was a suitable range for the strain to express higher inhibitory activities (Fig. 2C). The growth of strain B106 and its inhibitory activity against P. musae were detected in all of the tested media. The highest biomass (OD625 = 2.21) and inhibition rate (87.8%) were observed in EM. Moreover, the biomass and inhibition rates of strain B106 grown in PDB, BPYDB, Gause’s No. 1 and YDB media were higher than those in the other media except for EM. On the other hand, seven media including Czapek, Keshi No.1, PB, SS, ADS and SB were found to be suitable for the strain to express a stronger inhibitory activity against P. musae in which the inhibition rates were higher than 60%, but not for its growth (Fig. 2D). 3.4. Efficacy of strain B106 in controlling yellow Sigatoka in greenhouse

Fig. 3. Comparison of disease index (A) and control efficacy (B) among different treatments on suppressing the banana leaf spots caused by P. musae in greenhouse. The investigations were carried out 10 and 14 days after the treatments, respectively. The disease index (DI) and greenhouse control efficacy (GCE) was P calculated using the following formulae: DI ¼ ½100  ðNo: of affected leaves corresponding DSÞ=ðNo: of total leaves  9Þ; GCEð%Þ ¼ 100  ðDI of CK-DI of B106 or fungicide treatmentÞ=DI of CK. The disease severity (DS), expressed as a percentage of infected area over the total leaf area, was divided into five ratings. Values are the means of three replicates. Means with same letters are not significantly different according to Duncan’s multiple range test at P = 0.05. SA: strain B106 applied on the banana leaves 2 days after P. musae inoculation; SB: strain B106 applied on the banana leaves 2 days before P. musae inoculation; PA: propiconazole applied 2 days after P. musae inoculation.

The changes of DIs in the greenhouse trial are shown in Fig. 3A. The DI of the SB treatment (strain B106 applied on the banana leaves 2 days before P. musae inoculation) was 12.4 after 10 days

of application, and increased to 21.9 after 14 d days of application, which were lower than those of the SA treatment (strain B106 applied on the banana leaves 2 days after P. musae inoculation)

7

G. Fu et al. / Biological Control 55 (2010) 1–10

Table 2 Development of banana leaf spot diseases as represented by a disease index (DI) and as affected by applications of biocontrol strain B106 or propiconazole in 2007 and 2008 field trails. Disease index (DI)a

Year

Treatment

0 days

14 days

28 days

38 days

2007

Strain B106 Propiconazole CK

3.8ab 4.1a 3.4a

5.9ab 4.6c 6.1b

9.8b 6.4c 12.5b

17.8bc 10.7cd 28.1a

2008

Strain B106 Propiconazole CK

3.1a 2.6a 2.9a

7.2a 3.9c 7.8a

11.8b 5.6c 16.2a

20.2b 9.3d 36.6a

a The applications of strain B106/propiconazole were performed three times at 14-day-intervals. The investigations of disease index (DI) were carried out 14, 28 and P 38 days after the first application, respectively. The DI was calculated using the formula: DI ¼ ½100  ðNo: of affected leaves  corresponding DSÞ=ðNo: of total leaves  9Þ. The disease severity (DS), expressed as a percentage of infected area over the total leaf area, was divided into five ratings. b Values are the means of three replicates. Means in a column followed by same letters are not significantly different according to Duncan’s multiple range test at P = 0.05.

Table 3 Field control efficacy (FCE) of applications of biocontrol strain B106 or propiconazole against banana leaf spot diseases in 2007 and 2008 trials. Field control efficacy (FCE)a

Year

Treatment

14 days

28 days

38 days

2007

Strain B106 Propiconazole

11.2 bb 38.7b

29.5b 58.4b

41.2b 74.6a

2008

Strain B106 Propiconazole

15.6b 40.2a

33.2b 55.3a

48.3b 76.2a

a The applications of strain B106/propiconazole were performed three times at 14-day-intervals. The investigations of field control efficacy (FCE) were carried out 14, 28 and 38 days after the first application, respectively. The FCE was calculated using the formula: FCE (%) = [1  (BN  DI of treatment block at the 2nd or 3rd application)/(DI of treatment block at the 1st application  DI of CK block at the 2nd or 3rd application)]  100%. The disease index(DI) of the CK determined at the first application was used as the base number (BN). b Values are the means of three replicates. Means in a column followed by same letters are not significantly different according to Duncan’s multiple range test at P = 0.05.

and the CK, but higher than those of the PA treatment (propiconazole applied 2 days after P. musae inoculation). The greenhouse trial showed that 72.3% efficacy of strain B106 in controlling yellow Sigatoka caused by P. musae was observed 10 days after the antagonist application. The SB treatment was found to be more effective for suppressing subsequent infection by P. musae as compared to the SA treatment in which the antagonist application was done 2 days after inoculation with P. musae. The efficacy of strain B106 in controlling yellow Sigatoka was lower than that of propiconazole application. Significant differences in CE existed between strain 106 and propiconazole treatments or between each of the treatments and the CK. The efficacy of the antagonist and propiconazole in controlling the disease decreased after 14 days of incubation (Fig. 3B). 3.5. Efficacy of strain B106 in controlling the banana leaf spots naturally infected by multiple pathogens in the field Changes of the DIs of the banana leaf spots naturally infected by multiple pathogens in the field after the strain B106 and propiconazole applications in 2007 and 2008 are shown in Table 2. At the beginning of the trials, the DIs of the strain B106, propiconazole and CK were 3.4–4.1 for 2007 and 2.6–3.1 for 2008, respectively. No significant difference in DIs among the experimental plots was observed, indicating an even distribution of the leaf spot diseases in the field before spraying the antagonistic strain on the banana plants. After the 3rd application in 2007, the average DI of the strain B106 treatment was 17.8, which was significantly lower than CK (28.1), but higher than the propiconazole application (10.7). Similar results were observed in the second field trial in 2008. Efficacy of strain B106 in controlling the banana leaf spots in the field is shown in Table 3. In the first field trial in 2007, the CE increased with the increased numbers of the antagonist application

Fig. 4. Comparison of disease index (A) and control efficacy (B) between strain B106 and prochloraz treatments against banana post-harvest anthracnose disease. The investigations were carried out 14 and 17 days after the treatments, respectively. The disease index (DI) and control efficacy (CE) were calculated using following P formulae: DI ¼ ½100  ðNo: of diseased fruits  corresponding DSÞ=ðNo: of total fruits  9Þ; CE ð%Þ ¼ 100  ðDI of CK  DI of B106 or fungicide treatmentÞ=DI of CK. Values are the means of three replicates. Means with same letters are not significantly different according to Duncan’s multiple range test at P = 0.05.

and reached 41.2% 10 days after the last antagonist application. In the second field trial in 2008, strain B106 grown under the optimized condition showed higher CE as compared to the first one. The efficacy of the antagonist in controlling the leaf spots increased with the increased numbers of the antagonist application, and reached to 48.3% 10 days after its last application. The efficacy of

8

G. Fu et al. / Biological Control 55 (2010) 1–10

propiconazole in controlling the banana leaf spots were higher than that of the antagonist in both trials. Significant difference in CE existed between the fungicide and the antagonist treatments. 3.6. Efficacy of strain B106 in controlling banana post-harvest anthracnose The efficacy of strain B106 in controlling banana post-harvest anthracnose is shown in Fig. 4. After 14 days of the antagonist application, the average DI of the strain B106 treatment was 13.4 (Fig. 4A), which was significantly lower than that of CK (26.5), but higher than that of the prochloraz treatment (4.9). The average CE of the antagonist was 48.6% which was significantly lower than that of the prochloraz treatment (81.4%) at a significance level of P = 0.05 (Fig. 4B). After 17 days of storage, the average DI of the strain B106 treatment increased to 28.7 (Fig. 4A), which was substantially lower than that of CK (56.5), but higher than that of the prochloraz treatment (12.9). The average CE of the antagonist was 41.2% which was significantly lower than that of the prochloraz treatment (78.8%) at a significance level of P = 0.05 (Fig. 4B). 4. Discussion In this paper, the biocontrol strain B106 was identified using bacteriological and molecular methods. The morphological, physiological and biochemical characters of the antagonistic strain agreed with that of B. subtilis (Buchanan and Gibbons, 1984). The 16S rDNA sequence, an evolutionarily conserved sequence of bacteria, has been intensively used as a key reference for bacterial identification (Alvindia and Natsuaki, 2009; Ayyadurai et al., 2006; Zheng et al., 2008; Stackebrandt and Goebel, 1994), making the identification of a bacterium strain more fast and accurate. The results based on both bacteriological properties (Table 1) and 16S rDNA sequence analysis (Fig. 1) clearly demonstrated that the antagonist was a member of B. subtilis. As a biocontrol agent, Bacillus spp. has been reported for the control of single-pest-induced banana diseases, such as banana wilt (banana panama disease), banana anthracnose, banana root knot nematode, and banana bunchy top virus (Severn-Ellis, 2001; Chen et al., 2004; He et al., 2002; Janisiewicz and Korsten, 2002; Jonathan and Umamaheswari, 2006; Harish et al., 2009). Jiménez et al. (1987) reported Pseudomonas sp. as a biocontrol agent of black Sigatoka of bananas. However, the CE of Pseudomonas sp. was confirmed only under greenhouse conditions and its colonization ability found to be weak on the surface of banana leaves. Two strains of Serratia marcescens were also reported on controlling black Sigatoka of bananas in field with the same efficacy as conventional fungicides (González et al., 1996). However, the strain of Serratia marcescens isolated by Miranda could not control this leaf spot effectively (Miranda Corrales, 1996). This is the first report of B. subtilis as a biocontrol agent against the banana leaf spots caused by multiple pathogenic fungi in the field. The existence of an inhibition zone surrounding a colony of strain B106 indicated the production of unknown antibiotic substance(s) by the antagonist (data not shown). Induction of defence-related proteins by mixtures of plant growth promoting endophytic bacteria including Bacillus spp. against banana bunchy top virus has been reported (Harish et al., 2009). The possible mechanism(s) for strain B106 to suppress the banana leaf spot diseases could be considered as follows: (1) the production of the unidentified antibiotic substance(s) by the antagonist led to suppression of the banana leaf spots and (2) systemic acquired resistance of the banana plants might be induced by the antagonist. It is an important prerequisite to obtain a biocontrol agent with a higher environmental stress tolerance and a broad antagonistic

spectrum against multiple pathogens causing banana leaf spots for a successful field biocontrol of the diseases. The stress-tolerant ability of B. subtilis was found to be stronger than other bacteria lacking spores in their life cycles. The strain B106 had a broad antagonistic spectrum against all the prevalent fungal pathogens causing the banana leaf spots in Jinling township, Nanning city where the field trials were carried out (Fu et al., 2007). The experimental data in the present study suggested that the bacterial strain could be considered as a promising biocontrol agent for suppressing the banana leaf spot diseases regardless that its FCE was not as high as propiconazole, which was recognized as the most highly effective fungicide for controlling the banana leaf spots in the field. The initial inoculum level has been reported to affect growth of bacteria in media (Chen et al., 1998). In the present study, the initial concentrations of strain B106 in all of the media used for optimization of cultural conditions were at the same level, thereby enabling a comparison of biomass among various treatments. The experimental results indicated that the optimized cultural condition for strain B106 to express higher antagonistic activity against P. musae, the most prevalent pathogen causing banana leaf spots in the field, was the combination of 31 °C, pH 6.0, EM medium and 5day-incubation. However, the optimized cultural condition for the antagonist to produce higher biomass was the combination of 31– 34 °C, pH 6.5, EM medium and 3-day-incubation. Strain B106 was grown under the optimized cultural conditions in the second field trial. The antagonistic strain showed higher CE in the second field trial compared to the first one in which the bacterium was cultured under nonoptimized conditions (Table 3). The results suggested that the optimization of cultural conditions might be considered as a key point for the enhancement of FCE of the antagonist against the banana leaf spot diseases. In this work, only 16 kinds of liquid media were used for the optimization of cultural conditions. Potential for further optimization of cultural conditions still exists as the numbers of different media tested are expanded. Furthermore, improvements in field-applied conditions may also be another effective way. On the other hand, mutagenesis of the antagonist and screening of mutants with higher productivity of the antibiotic substance(s) might lead to further enhancement of its CE against the banana diseases. Purification and characterization of the antibiotic substance(s) produced by strain B106 is also a key task for better understanding of suppression of the banana diseases by the antagonist and for the development of new application approaches. In field trials, the CEs of the antagonist and propiconazole increased with the increased numbers of applications (Table 3). However, in the greenhouse (Fig. 3B) and post-harvest (Fig. 4B) trials, both the CEs of the antagonist and fungicide treatments decreased as time elapsed. The results indicated that successive applications of the antagonist or fungicide led to an escalation of CE. The decrease of CE in the one-time antagonist applications in both greenhouse and post-harvest trials suggested that the bacterium could not maintain its population on the leaves or fruits at a high level for a longer period. The reasons for the decrease of CE might be due to the nutrition deficiency on the surface of the banana leaves or fruits. On the other hand, disease suppression by strain B106 in the field was not as marked as those observed in the greenhouse. The reasons for the difference between GCE and FCE could be considered as follows: (1) The causal agent of the banana leaf spots in the greenhouse trial was a single pathogen (P. musae), while several pathogens including P. musae, C. musae, H. torulosa, A. musae, C. lunata, were involved in the banana leaf spots in the field trials. The sensitivities of these pathogens to the antagonist differed from each other (Fu et al., 2007). (2) The field environmental conditions might be more severe than those in greenhouse for the survival

G. Fu et al. / Biological Control 55 (2010) 1–10

and reproduction of the antagonist. The environmental stresses such as strong sunlight, dryness, high temperatures and competition with indigenous leaf microorganisms, were likely to reduce the colonization and biocontrol ability of strain B106 under field conditions. Therefore, further investigations on the population dynamics of the antagonist and other related ecological events on/in a banana plant might be necessary for the optimization of field-applied conditions in the future. Acknowledgments This research has been financed in part by grants from the Department of Science and Technology of Guangxi Zhuang Autonomous Region (Project Nos. 0728071 and 0991056), Science and Technology Bureau of Nanning City (Project No. 20060142B) and the special research grant from Guangxi Academy of Agricultural Sciences (Project No. 200925). References Alvindia, D.G., Kobayashi, T., Yaguchi, Y., Natsuaki, K.T., 2000. Symptoms and the associated fungi of postharvest diseases on non-chemical bananas imported from the Philippines. Japanese Journal of Tropical Agriculture 44, 87–93. Alvindia, D.G., Natsuaki, K.T., 2008. Evaluation of fungal epiphytes isolated from banana fruit surfaces for biocontrol of banana crown rot disease. Crop Protection 27, 1200–1207. Alvindia, D.G., Natsuaki, K.T., 2009. Biocontrol activities of Bacillus amyloliquefaciens DGA14 isolated from banana fruit surface against banana crown rot-causing pathogens. Crop Protection 28, 236–242. Amil, A.F., Heaney, S.P., Stanger, C., Shaw, M.W., 2007. Dynamics of QoI sensitivity in Mycosphaerella fijiensis in Costa Rica during 2000 to 2003. Phytopathology 97, 1451–1457. Anthony, S., Abeywickrama, K., Dayananda, R., WilsonWijeratnam, S., Arambewela, L., 2004. Fungal pathogens associated with banana fruit in Sri Lanka, and their treatment with essential oils. Mycopathologia 157, 91–97. Ayyadurai, N., Naik, P.R., Rao, M.S., Kumar, R.S., Samrat, S.K., Manohar, M., Sakthivel, N., 2006. Isolation and characterization of a novel banana rhizosphere bacterium as fungal antagonist and microbial adjuvant in micropropagation of banana. Journal of Applied Microbiology 100, 926–937. Buchanan, R.E., Gibbons, N.E., 1984. Bergey’s Manual of Determinative Bacteriology, Eighth ed. The Williams and Wilkins Co.Baltimore Md, Baltimore. Bureau, E., 1990. Adoption of a forecasting system to control black Sigatoka (Mycosphaerella fijiensis Morelet) in plantain plantations in Panama. Fruits 45, 329–338. Chen, C.Y., Wang, Y.H., Huang, C.J., 2004. Enhancement of the antifungal activity of Bacillus subtilis F29-3 by the chitinase encoded by Bacillus circulans chiA gene. Canadian Journal of Microbiology 50, 451–454. Chen, Z.Y., Lu, F., Liu, C.X., Liu, Y.F., Mew, T.H., 1998. Studies on culture condition and commercial production medium of antagonistic bacteria B. subtilis B-916 – a strain for biological control of sheath blight of rice. Southwest China Journal of Agricultural Sciences 12, 76–81. Chin, K.M., Wirz, M., Laird, D., 2001. Sensitivity of Mycosphaerella fijiensis from banana to trifloxystrobin. Plant Disease 85, 1264–1270. De Costa, D.M., Erabadupitiya, H.R.U.T., 2005. An integrated method to control postharvest diseases of banana using a member of the Burkholderia cepacia complex. Postharvest Biology and Technology 36, 31–39. de Lapeyre de Bellaire, L., Chillet, M., Mourichon, X., 2000. Elaboration of an early quantification method of quiescent infections of Colletotrichum musae on bananas. Plant Disease 84, 128–133. de Lapeyre de Bellaire, L., Dubois, C., 1997. Distribution of thiabendazole-resistant Colletotrichum musae isolates from Guadeloupe banana plantations. Plant Disease 81, 1378–1383. Demerutis, C., Quiros, L., Martinuz, A., Alvarado, E., Williams, R.N., Ellis, M.A., 2008. Evaluation of an organic treatment for post-harvest control of crown rot of banana. Ecological Engineering 34, 324–327. Fokkema, N.J., 1978. Fungal antagonism in the phyllosphere. Annals of Applied Biology 89, 115–117. Fu, G., Lin, S.H., Huang, S.L., Zhou, S.Q., Wang, Q., Cen, Z.L., 2007. Isolation and screening of antagonistic microorganisms against pathogens of banana leaf spot diseases in Guangxi. Southwest China Journal of Agricultural Sciences 20, 421– 424. González, R., Bustamante, E., Shannon, P., 1996. Evaluation de microorganismos quitinoliticos en el control de la Sigatoka negra (Mycosphaerella fijiensis) en banano. Manejo Integrado Plagas 40, 12–16. Harelimana, G., Lepoivre, P., Jijakli, H., Mourichon, X., 1997. Use of Mycosphaerella fijiensis toxins for the selection of banana cultivars resistant to black leaf streak. Euphytica 96, 125–128. Harish, S., Kavino, M., Kumar, N., Balasubramanian, P., Samiyappan, R., 2009. Induction of defense-related proteins by mixtures of plant growth promoting

9

endophytic bacteria against Banana bunchy top virus. Biological Control 51, 16– 25. Hayden, H.L., Carlier, J., Aitken, E.A.B., 2003. Population differentiation in the banana leaf spot pathogen Mycosphaerella musicola, examined at a global scale. Plant Pathology 52, 713–719. He, H., Cai, X.Q., Chen, Y.S., 2002. Biological control of banana anthracnose with endophytic Bacillus subtilis BS-2 and BS-1 isolated from Capsicum. Journal of Fujian Agricultural and Forestry University 31, 442–444. Huang, S.L., Kohmoto, K., 1991. A simple method for isolating single fungal spores. Bulletin of the Faculty of Agriculture, Tottori University 44, 1–3. Huang, S.L., Yan, B., Wei, J.G., Yan, W.H., Cen, Z.L., Yang, T., 2007. First report of plantain zonate leaf spot caused by Pestalotiopsis menezesiana in China. Australasian Plant Disease Notes 2, 61–62. Janisiewicz, W.J., Korsten, L., 2002. Biological control of postharvest diseases of fruits. Annual Review of Phytopathology 40, 411–441. ´ rez, C., 1987. Estudios sobre combate biologico de Jiménez, J.M., Galindo, J.J., Ramı Mycosphaerella fijiensis var. difformis mediante bacterias epifitas. In: Proceedings of the 7th Acorbat Meeting, Catie Turrialba, Costa Rica, pp. 105– 109. Jonathan, E.I., Umamaheswari, R., 2006. Bio-management of nematodes infesting banana by bacterial endophytes (Bacillus subtilis). Indian Journal of Nematology 36, 230–233. Jones, D.R., 2000. Introduction to banana, abaca and enset. In: Jones, D.R. (Ed.), Diseases of Banana, Abaca and Enset. CABI Publishing, Wallingford, UK, pp. 1– 36. Jones, D.R., 2003. The distribution and importance of the Mycosphaerella leaf spot diseases of banana. In: Jacome, L., Lepoivre, P., Marı´n, D., Ortiz, R., Romero, R., Escalant, J.V. (Eds.), Proceedings of the Second International Workshop on Mycosphaerella Leaf Spot Diseases, San José, Costa Rica, 20–23 May 2002, pp. 25–41. Khan, S.H., Aked, J., Magan, N., 2001. Control of the anthracnose pathogen of banana (Colletotrichum musae) using antioxidants alone and in combination with thiabendazole or imazalil. Plant Pathology 50, 601–608. Koné, D., Badou, O.J., Bomisso, E.L., Camara, B., Ake, S., 2009. Activités in vitro de différents fongicides sur la croissance chez Mycosphaerella fijiensis var. difformis Stover and Dickson, Cladosporium musae Morelet et Deightoniella torulosa (Syd.) Ellis, parasites isoles de la phyllosphére des bananiers en Côte-d’Ivoire. Comptes Rendus Biologies 332, 448–455. Koné, D., Ji, P., Fonsah, G.E., Csinos, A.S., 2008. First report of black leaf spot of banana caused by Deightoniella torulosa in Georgia. Plant Disease 92, 1470. Korsten, L., De Jager, E.S., De Villiers, E.E., Lourens, A., Kotzé, J.M., Wehner, F.C., 1995. Evaluation of bacterial epiphytes isolated from avocado leaf and fruit surfaces for biocontrol of avocado postharvest diseases. Plant Disease 79, 1149–1156. Krauss, U., Johanson, A., 2000. Recent advances in the control of crown rot of banana in the Windward Islands. Crop Protection 19, 151–160. Lassois, L., de Lapeyre de Bellaire, L., Jijakli, M.H., 2008. Biological control of crown rot of bananas with Pichia anomala strain K and Candida oleophila strain O. Biological Control 45, 410–418. Marín, D.H., Romero, R.A., Guzmán, M., Sutton, T.B., 2003. Black Sigatoka: an increasing threat to banana cultivation. Plant Disease 87, 208–222. Meredith, D.S., 1970. Banana leaf spot disease (Sigatoka) caused by Mycosphaerella musicola Leach. Phytopathological Papers, No. 11, Commonwealth Mycological Institute, Kew, Surrey, UK, pp. 147. Miranda Corrales, J.E., 1996. Evaluación de microorganismos antagonistas al hongo Mycosphaerella fijiensis (Morelet), colocados en el interior y exterior de la planta de banano. Catie Turrialba, Costa Rica. Mobambo, K.N., Gauhl, F., Swennen, R., Pasberg-Gauhl, C., 1996. Assessment of the cropping cycle effects of black leaf streak severity and yield decline of plantain and plantain hybrids. International Journal of Pest Management 42, 1–7. Mobambo, K.N., Gauhl, F., Vuylsteke, D., Ortiz, R., Pasberg-Gauhl, C., Swennen, R., 1993. Yield loss in plantain from black sigatoka leaf spot and field performance of resistant hybrids. Field Crops Research 35, 35–42. Niroshini Gunasinghe, W.K.R., Karunaratne, A.M., 2009. Interactions of Colletotrichum musae and Lasiodiplodia theobromae and their biocontrol by Pantoea agglomerans and Flavobacterium sp. in expression of crown rot of ‘‘Embul” banana. BioControl 54, 587–596. Pérez, L., Hernández, A., Hernández, L., Pérez, M., 2002. Effect of trifloxystrobin and azoxystrobin on the control of black Sigatoka (Mycosphaerella fijiensis Morelet) on banana and plantain. Crop Protection 21, 17–23. Ranasinghe, L., Jayawardena, B., Abeywickrama, K., 2002. Fungicidal activity of essential oils of Cinnamomum zeylanicum (L.) and Syzygium aromaticum (L.) Merr et L.M. Perry against crown rot and anthracnose pathogens isolated from banana. Letters in Applied Microbiology 35, 208–211. Ranasinghe, L., Jayawardena, B., Abeywickrama, K., 2005. An integrated strategy to control post-harvest decay of Embul banana by combining essential oils with modified atmosphere packaging. International Journal of Food Science and Technology 40, 97–103. Reyes, M.E.Q., Nishijima, W., Paull, R.E., 1998. Control of crown rot in ‘Santa Catarina Prata’ and ‘Williams’ banana with hot water treatments. Postharvest Biology and Technology 14, 71–75. Romero, C.R., 1986. Impacto de Sigatoka negra y roya del cafeto en actividad platanera nacional. Revista de la Asociación Bananera Nacional (ASBANA), San José, Costa Rica 12, 7–10. Romero, R.A., Sutton, T.B., 1997. Sensitivity of Mycosphaerella fijiensis, causal agent of black Sigatoka of banana, to propiconazole. Phytopathology 87, 96–100.

10

G. Fu et al. / Biological Control 55 (2010) 1–10

Romero, R.A., Sutton, T.B., 1998. Characterization of benomyl resistance in Mycosphaerella fijiensis, cause of black Sigatoka of banana, in Costa Rica. Plant Disease 82, 931–934. Severn-Ellis, A.A., 2001. Enhancement of tissue-cultured banana plant and root growth with Bacillus spp. In: Molina A.B., Nik Masdek N.H., Liew K.W. (Eds.), Banana Fusarium Wilt Management: Towards Sustainable Cultivation. International Network for the Improvement of Bananas and Plantains – Asia and the Pacific Network, Los Bahos, Laguna, pp. 275–281. Silva, M., Mendonca, H.L., Barreto, R.W., Pereira, O.L., 2008. First report of leaf spots on banana in Brazil caused by Bipolaris sacchari. Australasian Plant Disease Notes 3, 10–11. Stackebrandt, E., Goebel, B.M., 1994. Taxonomic note: a place for DNA–DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. International Journal of Systematic Bacteriology 44, 846–849. Stover, R.H., 1980. Sigatoka leaf spot of bananas and plantains. Plant Disease 64, 750–755. Surridge, A.K.J., Wehner, F.C., Crous, P.W., Viljoen, A., 2003. First report of Cladosporium musae on banana in South Africa. Australasian Plant Pathology 32, 499–503. Twizeyimana, M., Ojiambo, P.S., Tenkouano, A., Ikotun, T., Bandyopadhyay, R., 2007. Rapid screening of Musa species for resistance to black leaf streak using in vitro plantlets in tubes and detached leaves. Plant Disease 91, 308–314. Vawdrey, L.L., Peterson, R.A., DeMarchi, L., Grice, K.E., 2004. Evaluation of mineral oils and plant-derived spray adjuvants, mancozeb formulations and rates of application, for the control of yellow Sigatoka leaf spot (caused by

Mycosphaerella musicola) of bananas in far northern Queensland, Australia. Australasian Plant Pathology 33, 379–384. Washington, J.R., 1997. Relationship between the spray droplet density of two protectant fungicides and the germination of Mycosphaerella fijiensis ascospores on banana leaf surfaces. Pesticide Science 50, 233–239. Washington, J.R., Cruz, J., Lopez, F., Fajardo, M., 1998. Infection studies of Mycosphaerella fijiensis on banana and the control of black Sigatoka with chlorothalonil. Plant Disease 82, 1185–1190. Williams, C.M., Richter, C.S., Mackenzie Jr., J.M., Shih, J.C.H., 1990. Isolation, identification, and characterization of a feather-degrading bacterium. Applied and Environmental Microbiology 56, 1509–1515. Williamson, S.M., Guzmán, M., Marín, D.H., Anas, O., Jin, X., Sutton, T.B., 2008. Evaluation of Pseudomonas syringae strain ESC-11 for biocontrol of crown rot and anthracnose of banana. Biological Control 46, 279–286. Win, N.K.K., Jitareerat, P., Kanlayanarat, S., Sangchote, S., 2007. Effects of cinnamon extract, chitosan coating, hotwater treatment and their combinations on crown rot disease and quality of banana fruit. Postharvest Biology and Technology 45, 333–340. Zheng, Y.G., Chen, J., Liu, Z.Q., Wu, M.H., Xing, L.Y., Shen, Y.C., 2008. Isolation, identification and characterization of Bacillus subtilis ZJB-063, a versatile nitrite-converting bacterium. Applied Microbiology and Biotechnology 77, 985–993. Zimmerman, A., 1902. Ueber einig´e tropischer kulturpfanzen beobachtete pilze. Zentralblatt für Bakteriolog´ie, Parasitenkunde, Infektionskrankbeiten und Hyg´iene 8, 219.