Phân lập và tuyển chọn các chủng vi khuẩn sinh enzyme chitinase, đối kháng với nấm Fusarium oxysporum từ đất vùng rễ cây hoa lily

Vietnam J. Agri. Sci. 2020, Vol. 18, No.11: 965-975 Tạp chí Khoa học Nông nghiệp Việt Nam 2020, 18(11): 965-975 www.vnua.edu.vn 965 ISOLATION AND SELECTION OF CHITINASE-PRODUCING BACTERIA WITH ANTIFUNGAL ACTIVITY AGAINST FUSARIUM OXYSPORUM FROM LILIUM RHIZOSPHERE SOIL Tran Thi Hong Hanh*, Tran Thi Dao, Le Thi Ngoc An, Lam Thuong Thuong Department of Microbial Biotechnology, Faculty of Biotechnology, Vietnam National University of Agriculture *Email: honghanh@vnua.edu.vn Received: 03.07.2018 Published: 25.06.2020 ABSTRACT Chitinases have many applications in various fields such as environment, agriculture, and medicine. Chitinolytic bacteria, which are found in various natural environments including the rhizosphere, soil, marine ecosystems, lakes, and chitinous waste, can be used for the degradation of chitin-containing wastes as well as in the production of chitinolytic enzymes with fungicidal activity against some fungal phytopathogens. This study consisted of isolating and screening chitinolytic bacteria based on a chitinase activity assay using the diffusion plate and colorimetric methods. Thirty-two chitinase-producing bacterial strains were isolated from Lilium rhizosphere soil. There were four strains, namely HP02, VRQ9, HM03, and GL10, which exhibited the highest chitinase activities, with values of 1.47, 1.44, 1.29, and 1.21 U/ml, respectively. These chitinase-producing bacterial strains were tested for antagonistic activity against the causal agent of bulb and stem rots of lilies, Fusarium oxysporum, and the results indicated antifungal activity of the four strains HP02, VRQ9, HM03, and GL10. The HP02 strain had the highest level of chitinase activity and was capable of being antagonistic to Fusarium oxysporum. Based on morphological and biochemical characteristics, 16S rRNA gene sequencing, and phylogenetic analysis of this sequence along with sequences from GenBank, HP02 was identified as Bacillus subtilis. Keywords: Chitinase-producing bacteria, Fusarium oxysporum, rhizosphere. Phân lập và tuyển chọn các chủng vi khuẩn sinh enzyme chitinase, đối kháng với nấm Fusarium oxysporum từ đất vùng rễ cây hoa lily TÓM TẮT Enzyme chitinase có tiềm năng ứng dụng trong nhiều lĩnh vực như môi trường, nông nghiệp và y học. Vi khuẩn sinh enzyme chitinase được tìm thấy trong môi trường tự nhiên (đất vùng rễ, nước biển) được ứng dụng để đối kháng với nấm gây bệnh thực vật và sản xuất enzyme chitinase phân huỷ các chất thải chứa chitin. Nghiên cứu đã tiến hành phân lập và tuyển chọn các chủng vi khuẩn có khả năng sinh enzyme chitinase có hoạt tính cao bằng phương pháp khuyếch tán đĩa thạch và phương pháp so màu với thuốc thử DNS (3,5-dinitrosalicylic acid). Từ đất vùng rễ cây hoa lily, 32 chủng vi khuẩn có khả năng sinh enzyme chitinase đã được phân lập. Trong đó, 4 chủng HP02, VRQ9, HM03 và GL10 có hoạt tính chitinase là cao nhất, lần lượt là 1,47; 1,44; 1,29; 1,21 U/ml. Các chủng sinh chitinase được thử đối kháng với nấm Fusarium oxysporum gây bệnh thối củ và thân ở cây hoa lily. Kết quả cho thấy rằng 4 chủng HP02, VRQ9, HM03 và GL10 đều có khả năng đối kháng với nấm. Chủng HP02 có hoạt tính chitinase cao nhất và có khả năng đối kháng nấm Fusarium oxysporum, được xác định là Bacillus subtilis dựa vào đặc điểm hình thái, hoá sinh, so sánh trình tự 16S rRNA với các trình tự trong ngân hàng gen và xây dựng câyphân loại. Từ khoá: Vi khuẩn sinh chitinase, Fusarium oxysporum, vùng rễ. 1. INTRODUCTION Chitinases are glycosyl hydrolases catalyzing the degradation of chitin, which is a homopolymer of beta-1,4-linked N-acetyl-D glucosamine residues. In nature, chitin is the second most abundant polysaccharide, next to cellulose. Chitinases have many applications Isolation and selection of chitinase-producing bacteria with antifungal activity against Fusarium oxysporum from Lilium rhizosphere soil 966 such as a biocontrol agent of fungal diseases of plants in agriculture fields, the degradation of wastes containing chitin in environmental protection, the preparation of pharmaceutically important chitooligosaccharides and N-acetyl D-glucosamine in medicine, and the generation of protoplasts from fungi and yeasts (Neeraja et al., 2010; Hamid et al., 2012). These enzymes can be found in a wide range of organisms such as bacteria, fungi, insects, animals, and plants (Hoster et al., 2005). Some of the best known chitinase-producing bacteria belong to the Aeromonas, Serratia, Streptomyces, and Bacillus genera (Jabeen and Qazi, 2014). Chitinase-producing bacteria not only degrade chitin and utilize it as an energy source but also control plant diseases caused by various phytopathogenic fungi such as Fusarium oxysporum (the causal agent of bulb rot of lilies) (Lim et al., 2003), Verticillium dahliae (the causal agent of Verticillium wilt of cotton) (Li et al., 2013), Phaeoisariopsis personata (the causal agent of late leaf spot disease of groundnut) (Kishore et al., 2005a), Aspergillus niger (the causal agent of collar rot) (Kishore et al., 2005b), and Pyricularia oryzae/Magnaporthe oryzae (the causal agent of rice blast disease) (Nguyen et al., 2013). Chitinases inhibit fungal growth by hydrolyzing the chitin present in the fungal cell wall (Hoster et al., 2005). Therefore, chitinase-producing bacteria have promise as replacements for the application of synthetic pesticides and fungicides (Veliz et al., 2017). Chitinolytic bacteria can be found in many different environments such as soil, marine ecosystems, lakes, or chitinous waste (Setia & Suharjono, 2015). In soil, bacteria show vast variation quantitatively and qualitatively across different collection sites and at various depths (Mukerji et al., 2006). Soil bacteria are the major source of chitinase production and could be used for bioconversion of chitinous waste into useful molecules and for applications in different areas such as medicine, biotechnology, and agriculture (Jabeen & Qazi, 2014). The rhizosphere region is the portion of the soil which is adjacent to the root system of a plant and is influenced by root exudates such as amino acids, vitamins, sugars, and tannins (Mukerji et al., 2006). It is the first line of defense for roots against attacks by pathogenic fungi (Chet et al., 1990). Therefore, microorganisms growing in the rhizosphere region have the potential to be used as biocontrol agents (Podile and Prakash, 1996). The rhizosphere is populated by different microorganisms including both beneficial and deleterious ones (Chet et al., 1990), and there is a need to learn more about the beneficial rhizosphere microorganisms. Chitinolytic microorganisms, especially bacteria and actinomycetes, are heavily colonized in soil and the rhizosphere. In comparison with chitinolytic bacteria isolated from water and bottom sediments, chitinolytic bacteria isolated from soil have been shown to usually be more active (Brzezinska et al., 2014). Therefore, they could be more appropriate for agricultural use. Lily is one of the most important bulbous plants worldwide for production as cut flowers, for potted plant purposes, and for landscape decoration. The genus Lilium, which is a perfect target for exploitation by the ornamental industry, includes more than 80 species with significant differences in flower shapes, colors, sizes, fragrances, and bulb morphologies (Nunez de Caceres Gonzalez et al., 2015). The soil- borne fungus Fusarium oxysporum is a serious threat to the bulb and flower production of members of the Lilium genus (Straathof & Van- Tuyl, 1994). The objectives of this research were to isolate and determine potential bacteria from Lilium rhizosphere soil with high chitinase production, and to screen them for their antagonistic activities against Fusarium oxysporum for managing bulb and stem rots of lilies. 2. MATERIALS AND METHODS 2.1. Soil sampling Seven rhizosphere soil samples were collected at different Lilium fields with the help of sterile spatulas in various locations in Ha Tran Thi Hong Hanh, Tran Thi Dao, Le Thi Ngoc An, Lam Thuong Thuong 967 Noi, Hai Phong, Bac Ninh, and Ninh Binh. These samples were kept in plastic bags and brought to the laboratory for preservation at 4oC until isolation (Amin et al., 2015). 2.2. Isolation and screening of chitinase- producing bacterial strains To isolate the bacteria, the soil samples were diluted by the serial dilution method. One gram of a Lilium rhizosphere soil sample was added into 10ml sterile water and mixed thoroughly by vortex. The initial solution (10-1) was subsequently diluted to a 10-8 dilution. One-hundred microliters of the dilutions from 10-5 to 10-8 were spread on agar plates with Luria Bertani (LB) medium containing the following components: 10.0 g/l peptone; 5.0 g/l NaCl; 5.0 g/l yeast extract; and 20.0 g/l agar (pH 7.0). The plates were incubated at 37C for 3 days to isolate single bacterial colonies. The obtained colonies were then purified by the three phase streak technique. The pure bacterial colonies were cultured in liquid LB media supplemented with colloidal chitin 1% (w/v) and incubated at 37oC for 48 hours on a rotary shaker (180 rpm). After incubation for 48 hours, the culture broth was centrifuged at 10,000 rpm for 10 min and the supernatants were collected for measurement of chitinolytic activities (Sadfi et al., 2001). The isolates were transferred to LB agar slants for storage at 4C and sub-cultured once a month or maintained at -20C in LB medium containing 30% glycerol for long-term storage (Daamin-Remadi et al., 2006). 2.3. Enzyme assay 2.3.1. Preparation of colloidal chitin The colloidal chitin was prepared by using 1 g of chitin powder added to H3PO4 85% and kept for 24 hours at 4C. The colloidal chitin prepared for each of the samples was washed several times with a large volume of distilled water to adjust the pH to 7.0 (Alam & Mathur, 2014). 2.3.2. Detection of chitinolytic activity on plates Chitinolytic activities of the bacterial strains were determined by the diffusion plate method (Dhanasekaran et al., 2012). The sterilized chitin agar medium containing 1.0g colloidal chitin, 2.0g agar, and 100 ml phosphate buffer (pH 7.0) was prepared in Petri dishes. Four wells were made in each plate using a sterile well cutter. Fifty microliters of each chitinase-containing supernatant of the different strains was added into each well in the plates. The plates were placed in a refrigerator at 4C for 2 hours and then incubated at 30C for 24 hours. After incubation, Lugol solution was added to the plates for observing and recording the clear zone around the wells, indicating that the hydrolysis of chitin had occurred. Each assay was carried out three times and mean values were calculated. Chitinase activity was identified by the following formula: K = D – d, where, K is the chitinase activity, D is the diameter of the clear zone, and d is the diameter of the well (d = 7mm). 2.3.4. Determination of chitinolytic activity based on colorimetric method The chitinolytic activity of each isolate was determined by the estimation of released reducing sugars from the chitin. The mixture of 1ml of cell free supernatant and 1ml of 1% colloidal chitin suspension was incubated at 40C for 1 hour. The reaction was terminated by adding 1ml 1N NaOH and boiling the mixture at 100C for 5 min. After centrifugation of the reaction mixture at 10,000 rpm for 10 min, 1 ml of solution containing reducing sugars (N- acetylglucosamine, NAG) was added by 1ml DNS and heated at 100C for 10 min in a boiling water bath. After cooling to room temperature, the solution was measured for OD at  = 535nm. The amount of reducing sugars released was calculated from the standard curve for NAG (Sadfi et al., 2001; Jabeen & Qazi, Isolation and selection of chitinase-producing bacteria with antifungal activity against Fusarium oxysporum from Lilium rhizosphere soil 968 2014). The experiment was performed in triplicate and results represented mean values. 2.4. Antifungal assay For the detection of antifungal activity of the chitinase-producing bacteria, the inhibition zone assay was carried out. The mycelium of the phytopathogenic fungus Fusarium oxysporum was inoculated in the center of agar plates containing potato dextrose agar (PDA) medium (20.0 g potato, 20.0g D-glucose, 20.0g agar, and 1.0 liter distilled water, pH 7.0) and incubated at 30C for three days. This fungus was provided by the Department of Microbial Biotechnology, Faculty of Biotechnology, Vietnam National University of Agriculture. When the radial diameter of the fungal colonies was about 2cm, 50l of the supernatants from the four strains with high chitinolytic activities were loaded in wells located equidistantly around the center of the plates and then incubated at 30C for three days. The diameters of the inhibition zones were measured. The experiment was repeated three times (Et- Katatny et al., 2000). 2.5. Morphological and biochemical analysis The bacterial strain that had the highest chitinolytic activity and was capable of inhibiting the growth of the phytopathogenic fungus Fusarim oxysproum was selected and characterized further through tests, namely, the Gram stain, catalase, motility, citrate, sugar utilization, and Voges-Proskauer (VP) tests, according to the methods described by Barrow & Feltham (2004). 2.6. Identification of isolates Identification of the isolated bacteria was carried out by 16S rRNA sequence analysis. The 16S rRNA genes of the isolated bacteria were amplified by PCR using the forward primer 27F (5’-AGATTTGATCCTGGCTCAG- 3’) and the reverse primer 1492R (5’- GGTTACCTT GTTACGACTT-3’). The final volume for the PCR amplification was 25l including 1l primer 27F, 1l primer 1492R, 2l DNA template, 8.5l H2O, and 12.5l Mastermix 2X. The program for PCR amplification was performed using a Bio-Rad Thermal Cycler C1000 as follows: an initial denaturation at 94C (3 min) followed by 30 cycles of denaturation at 94C (1 min), annealing at 57C (2 min), and extension at 72C (2 min), and a final extension at 72C (4 min). The PCR products were visualized using gel electrophoresis on 1% agarose. Then, 5l of PCR products were sent to 1st BASE (Singapore) for DNA sequencing in an automated DNA sequencer. The sequence was compared to the GenBank database available at NCBI using the BLAST search engine for recognition of the highest percent similarity with the described strains and aligned through the ClustalW Multiple Alignment tool. A phylogenetic tree was constructed by the neighbor-joining method using MEGA 6.0 with 1000 bootstraps (Hoster et al., 2005; Setia and Suharjono, 2015). 3. RESULTS AND DISCUSSION 3.1. Isolation and screening of chitinase- producing bacterial strains From seven Lilium rhizosphere soil samples, a total of 61 bacterial strains were isolated. The results consisted of 42 strains from Ha Noi (accounting for the highest percentage of 68.9 %), 4 strains from Hai Phong (6.6%), 3 strains from Bac Ninh (4.9%), and 12 strains from Ninh Binh (19.6%). These strains were labeled as follows: GL01 - GL16; BT01 - BT13; HM01 - HM03; VRQ1 - VRQ10; HP01 - HP04; BN01 - BN03; and NB1 - NB12. The colonies of the isolated bacterial strains exhibited different morphological characteristics in their size, shape, surface, margins, elevation, color, and texture. Their sizes ranged from 1 to 5mm. The colonies of the isolates were circular or irregular in shape, had entire or undulate margins, were raised or flat in elevation, appeared yellowish, orange, white or cream in color, and appeared dry and Tran Thi Hong Hanh, Tran Thi Dao, Le Thi Ngoc An, Lam Thuong Thuong 969 wrinkled or shiny and smooth in texture. Most of them had a circular shape, flat elevation, and dry texture. Using a light microscope, the cellular morphologies of these isolated strains were rod-shaped, occurring in singles or pairs. All of the isolated strains were screened for chitinase activity based on their zone of clearance on colloidal chitin agar plates as described above. Out of the 61 different isolates tested, 31 bacterial strains were identified as chitinase producers based on their clear zone production, accounting for 52.2%. The zones of clearance surrounding the wells indicated chitinase activity capable of breaking down the chitin compound in the medium. These results showed that Lilium rhizosphere soil was quite abundant in the number of chitinase-producing bacteria. The chitinase activities of some of the isolated bacterial strains are shown in Table 1 and Figure 1. As described in Table 1, only four bacterial isolates (HP02, VRQ9, GL10, and HM03) were capable of producing chitinase with high chitinolytic activity (k >20mm), and had zones of 21.17, 20.83, 20.50, and 20.12mm, respectively. These four strains were the best producers of chitinase, and were chosen for determining their exact chitinase activities according to the colorimetric method. Table 1. Screening of bacterial isolates for chitinase activities No. Strain Chitinase activity (k, mm) No. Strain Chitinase activity (k, mm) 1 GL3 15.33 ± 0.58 10 VRQ6 13.83 ± 0.76 2 GL4 17.17 ± 0.76 11 VRQ9 20.83 ± 0.29 3 GL10 20.50 ± 0. 50 12 HP02 21.17 ± 0.29 4 GL11 7.83 ± 0.29 13 HP03 5.17 ± 0.29 5 BT02 16.33 ± 0.58 14 BN03 16.50 ± 0.50 6 BT13 15.67 ± 0.76 15 NB1 16.12 ± 0.58 7 HM03 20.12 ± 0.29 16 NB2 15.83 ± 0.29 8 VRQ3 13.67 ± 0.58 17 NB3 15.50 ± 0.50 9 VRQ5 13.67 ± 0.58 18 NB9 15.33 ± 0.58 Notes: k <15mm: weak chitinase activity; 15mm k <20mm: medium chitinase activity; 20mm k <25mm: strong chitinase activity; k 25mm: very strong chitinase activity. Figure 1. Chitinase activities of a selction of the isolated strains based on their clear zones around the wells after 2 days incubation Isolation and selection of chitinase-producing bacteria with antifungal activity against Fusarium oxysporum from Lilium rhizosphere soil 970 Figure 2. Chitinase activities of the four selected bacterial strains based on measurements of the release of reducing sugars (NAG) Figure 3. Antagonistic effects of some isolated bacterial strains against Fusarium oxysporum Tran Thi Hong Hanh, Tran Thi Dao, Le Thi Ngoc An, Lam Thuong Thuong 971 3.2. Chitinase activity of the four selected strains based on the colorimetric method The chitinase enzymes secreted by the four isolated bacterial strains were evaluated using the colorimetric assay and their chitinase activities are illustrated in Figure 2. Among the four strains, the HP02 strain gave the highest level of chitinase activity, with a maximum enzyme production of 1.47 ± 0.02 (U/ml) after 2 days incubation, while HM03 exhibited the lowest chitinolytic activity with 1.21 ± 0.06 (U/ml). Thus, the HP02 strain was selected for morphological and biochemical characterization and molecular identification based on 16S rRNA gene sequencing. Trachuck et al. (1996) reported the chitinase activities of different segregant-type colonies, namely, rough (R), smooth (S), and mucous (M) colonies of Bacillus licheniformis B- 6839. The authors observed that these variants (R, S, M) showed marked differences in chitinase activity levels, measuring 0.8, 1.3, and 1.7 U/ml, respectively. In 2007, Kamil et al. isolated four isolates (MS1, MS2, MS3, and MS4), which were the most potent chitinolytic bacteria from Egyptian rhizosphere soil samples of three different plants (maize, wheat, and rice). The MS3 strain, which was identified as Bacillus licheniformis, produced the highest level of chitinase enzymes (1.27 U/ml). However, the chitinolytic activity of the MS3 strain was still lower than that of the HP02, VRQ9, and GL10 strains, which recorded 1.47, 1.44, and 1.29 U/ml, respectively. Due to the high levels of chitinase production, these strains are considered to be potential biocontrol agents of plant diseases (Chang et al., 2003; Kamil et al., 2007) and were chosen for testing their antifungal activities against Fusarium oxysporum. 3.3. Antifungal activity of isolated strains Screening tests for antagonism of the isolated strains against Fusarium oxysporum were carried out as described above. The results of the inhibitory activities against Fusarium oxysporum are shown in Figure 3. It was apparent that four strains (HP02, VRQ9, GL10, and HM03) had inhibitory activities against this phytopathogenic fungus. The radial growth of the plant pathogenic fungus (Fusarium oxysporum) was restricted and inhibition zones around these bacterial strains were observed. These observations might be linked to the production of chitinase or other antibiotics. Bacterial chitinases can effectively hydrolyze the cell walls of phytophathogenic fungi due to the disruption of the glycosidic bonds in the chitin which serve as a fibrous-strengthening element in the fungal cell wall. As can be seen from Figure 3, the four isolates (HP02, VRQ9, GL10, and HM03) had high chitinolytic activities resulting in significantly higher antifungal activities than all the other isolated strains tested. Therefore, they were determined to have the potential efficacy for application as biocontrol agents. The biocontrol activity against plant pathogens of rhizobacteria is not only because they produce hydrolytic enzymes (chitinase and glucanase) but also because they synthesize other antimicrobial compounds such as siderophores and antibiotics (Ahmad et al., 2017). Interestingly, although the HP02 strain had the highest chitinase activity, the antifungal activity of this strain was lower than that of the isolated VRQ9 strain. Agrobacterium, Alcaligenes, Arthrobacter, Bacillus, Cellulomonas, Enterobacter, Pseudomonas, and Serratia are genera that are ideal for biocontrol (Podile and Prakash, 1996). Pseudomonas and Bacillus species living in the rhizosphere have been used to control diseases caused by a variety of plant pathogens including Fusarium spp. (Nam et al., 2009). Nam et al. indicated that Bacillus subtilis BS87 and RK1 were capable of suppressing the mycelial growth of Fusarium oxysporum f. sp. fragariae (the causal agent of Fusarium wilt in strawberry). Strain RK1 exhibited the highest level of fungal growth inhibition. Bacilli are the most popular bacterial biocontrol agents Isolation and selection of chitinase-producing bacteria with antifungal activity against Fusarium oxysporum from Lilium rhizosphere soil 972 because they are able to form endospores that are tolerant to heat, desiccation, and organic solvents. This is an advantage over some root colonizing bacteria. Besides Fussarium spp., Bacillus subtilis has also been shown to be inhibitory to other fungal pathogens such as Aspergillus niger (Podile & Prakash, 1996). The results obtained by Kamil (2017) indicated that Bacillus licheniformis strain MS3 was the most active species leading to the suppressed growth of various tested pathogenic fungi, namely, Rhizoctonia solani, Macrophomina phasiolina, Fusarium culmorum, Alternaria alternate, and Sclerotium rolfsii. The three bacterial strains HP02, VRQ9, and GL10, which had higher levels of chitinase activities than that of MS3, might be resistant to Fusarium oxysporum. Therefore, they might have the potential to have antagonistic activity against other fungal pathogens. Kishore et al. (2005b) indicated that Pseudomonas sp. GRS175 and Pseudomonas aeruginoase GPS 21 were highly inhibitory against fungal pathogens of groundnut. Pseudomonas fluorescens MM-B16 had protective activities against Phytophthora disease of pepper and anthracnose of cucumber (Lee et al., 2003). Pseudomonas and Bacillus are also important members of plant-growth promoting bacteria (PGPB), which induce plants to tolerate environmental stresses and can be used as biofertilizers (Ahmad et al., 2017). Mishara et al. (2012) observed that the Bacillus subtilis strain WR-W2, which has high similarity with the HP02 strain (99%), was found to be capable of synthesizing siderophore and indole acetic acid (IAA). Bacillus amyloliquefaciens FZB42, which was isolated from the rhizosphere of Beta vulgaris (sugar beet), Bacillus amyloliquefaciens NJN-6 (from the banana rhizosphere), and Bacillus subtilis HJ5 (from the cotton rhizosphere) had plant growth promoting activity and disease suppression based on their abilities for plant root colonization and antibiotic production (Li et al., 2013; Chowdhury et al., 2015). 3.4. Morphological and biochemical characteristics of HP02 The morphological and biochemical characteristics of HP02 strain are shown in Figure 4 and Table 2. HP02 was found to be Gram positive, with spore-forming rod bacterium, and formed colonies that were 3-5 mm in size with circular form, undulate margins, flat elevation, dry in texture, and cream in color. Cells of this strain were short rod-shaped, occurring in singles or in pairs. Because the endospores strongly resisted the stain dyes, they appeared as non- staining entities in the Gram stain. As indicated in Figure 4(b), endospores were extremely refractive. The isolated strain HP02 was catalase-positive, aerobic, and motile, and had distinct fermentation profiles for the carbon sources maltose, xylose, lactose, mannitol, and glucose. (a) (b) Note: (a) Colony of the HP02 strain; (b) Cellular morphology of the HP02 strain (magnification x 1000). Figure 4. Morphological characteristics of the HP02 strain Tran Thi Hong Hanh, Tran Thi Dao, Le Thi Ngoc An, Lam Thuong Thuong 973 Table 2. Morphological and biochemical tests for the identification of the HP02 strain Tests Morphological/Biochemical characterization Gram staining + Shape Rod Spore formation + Motility + Catalase + Citrate utilization + Voges-Proskauer (VP) + Fermentation for different carbon sources Maltose + Xylose + Lactose + Mannitol + Glucose + Based on the observed phenotypic and biochemical characteristics, the HP02 strain was grouped into the genus Bacillus according to the description in Bergey’s Manual of Determinative Bacteriology (Holt et al., 1994). 3.5. Phylogenetic analysis of 16S rRNA sequences Phylogenetic analysis of the 16S rRNA gene sequence using MEGA6 revealed that the isolated HP02 strain was Bacillus subtilis (Figure 5). The sequence of the HP02 strain and closely related sequences were analyzed using the neighbor-joining method. The GenBank accession number of each isolate is given in the parentheses. Bootstrap values based on 1000 replicates are shown next to the branches. This strain showed high sequence similarity (99%) to Bacillus subtilis strain WR-W2 (FJ222553), and 98% to the closest known species in the Genbank database including Bacillus subtilis strain VS15 (KT210118), Bacillus subtilis strain CICC10090 (AY971357), and Bacillus subtilis strain PNS-29 (JQ218450). The phylogenetic tree showed that the HP02 strain is located in the same branch with Bacillus subtilis strain VS15 with a bootstrap value of 100. Sequencing using the 16S rRNA gene is a good method for resolving problematic phenotypic identifications and placing isolated bacteria in their right taxonomic position (Gomaa and Momtaz, 2006). The HP02 strain, which was identified as Bacillus subtilis, was a chitinolytic bacterial strain. Members of the genus Bacillus including Bacillus amyloliquefaciens, Bacillus cereus, Bacillus subtilis, Bacillus licheniformis, Bacillus laterosporus, Bacillus magaterium, Bacillus pabuli, and Bacillus thuringiensis have been reported to secrete a number of degradative enzymes such as chitinase (Shanmugaiah et al., 2008; Karunya et al., 2011). The results of this study also confirm the same results. 4. CONCLUSIONS Chitinases have many applications in different fields such as waste management, pest and pathogen control in agriculture, the generation of fungal protoplasts, and human health care. The present study indicated that four bacterial strains (HP02, VRQ9, HM03, and GL10) isolated from Lilium rhizosphere soil not only produced chitinase with high chitinolytic activity but also were capable of controlling Fusarium oxysporum (the causal agent of bulb and stem rots of lilies). Among them, the HP02 strain showed the highest chitinolytic activity. Based on morphological and biochemical characteristics, and 16S rRNA gene sequencing, HP02 was identified as Bacillus subtilis. This isolated strain is considered as a potential chitinase producer and as a biological control agent for the phytopathogenic fungus Fusarium oxysporum. Isolation and selection of chitinase-producing bacteria with antifungal activity against Fusarium oxysporum from Lilium rhizosphere soil 974 Note: The tree was constructed by the neighbor-joining method using MEGA6 software. Nucleotide sequence accession numbers are indicated in the parentheses. Figure 5. Phylogenetic tree representing the HP02 strain under study compared to selected bacteria from the GenBank database ACKNOWLEDGEMENTS The authors would like to thank the Department of Microbial Biotechnology, Faculty of Biotechnology for providing us with the phytopathogenic fungus Fusarium oxysporum. We are thankful for financial support from the Belgo-Vietnamese Project. REFERENCES Admad Z., Wu J., Chen L. & Dong W. (2017). Isolated Bacillus subtilis strain 330-2 and its antagonistic genes identified by the removing PCR. Sci. Rep. 7: 1-13. Alam J. & Mathur A.(2014). Evaluation of antifungal potential of chitin and chitin - based derivatives against pathogenic fungal strains. Biolife. 2(4): 1354-1358. Amin M., Rakhisi Z. & Ahmady A.Z. (2015). Isolation and identification of Bacillus species from soil and evaluation of their antibacterial properties. Avicenna Clin. Microb. Infec. 2(1): 1-4. Barrow G.I. & Feltham R.K.A. (2004). Cowan and Steel’s Manual for the identification of medical bacterial. 3rd ed., Cambridge University Press. pp. 217-232. Brzezinska M.S., Jankiewicz U., Burkowska A. & Walczak M. (2014). Chitinolytic microorganisms and their possible application in environmental protection. Curr. Microbiol. 68: 71-81. Chang W.T., Chen C.S. & Wang S.L. (2003). An antifungal chitinase produced by Bacillus cereus with shrimp and crab shell powder as a carbon sources. Curr. Microbiol. 47: 102-108. Chet I., Ordentlich A., Shapira R. & Oppenheim A. (1990). Mechanisms of biocontrol of soil-borne plant pathogens by rhizobacteria. Plant Soil. 129: 85-92. Chowdhury S. P., Hartmann A., Gao X. & Borriss R. (2015). Biocontrol mechanism by root-associated Bacillus amyloliquefaciens FZB42 - a review. Front. Microbiol. 6: 1-11. Daami-Remadi M., Ayed F. & Jabnoun-Khiareddine H. (2006). Effects of some Bacillus sp. isolates on Fusarium spp. in vitro and potato tuber dry rot development in vivo. Plant Pathol. 5(3): 283-290. Bacillus subtilis strain WR-W2 (FJ222553) Bacillus subtilis strain PNS-29 (JQ218450) Bacillus subtilis strain H13 (KT907045) Bacillus subtilis strain MB13 (HM565142) Bacillus subtilis subsp. spizizenii strain QAU63 (KC679985) Bacillus subtilis strain HS2 (KP743119) Bacillus subtilis CICC10148 (AY971358) HP02 Bacillus subtilis strain VS 15 (KT210118) Bacillus subtilis strain CTB04 (KF414530) Bacillus amyloliquefaciens strain PVSEB2 (KY741854) Bacterium strain strain Te25R (AY587811) Bacillus subtilis strain CICC10090 (AY971357) Bacillus subtilis strain GCUF-S1 (KM282517) Bacillus atrophaeus strain km51 (JF411244) Bacillus licheniformis strain G (MF871790) Bacillus sonorensis strain 3SSSS (KF574385) Bacillus licheniformis strain CMB25 (HM100196) Bacillus safensis strain BL-01 (KX082859) Bacillus safensis strain RD3 (KJ621590) Bacillus pumilus strain SWS09 (HM100218) Bacillus pumilus strain 28UU (KF574388) Bacillus stratosphericus strain MSRK BC 4 (KY485271) Bacillus firmus strain MB5 (JN215489) Bacillus aquimaris strain RM-25 (KU352785) Bacillus cereus (AJ277907) 10 92 82 80 74 97 68 91 80 66 99 60 59 53 100 51 80 36 97 90 92 26 45 Tran Thi Hong Hanh, Tran Thi Dao, Le Thi Ngoc An, Lam Thuong Thuong 975 Dhanasekaran D., Thajuddin N. & Panneerselvam A. (2012). Applications of Actinobacterial Fungicides in Agriculture and Medicine. Fungicides for Plant and Animal Diseases. Khanasekaran D. (Ed.). INTECH. pp. 29-54. El-Katatny M.H., Somitsch W., Robra K.H., El- Katatny M.S. & Gubitz G.M. (2000). Production of chitinase and β- 1, 3 - glucanase by Trichoderma harzianum for control of the phytopathogenic fungus Sclerotium rolfsii. Food Technol. Biotechnol. 38(3): 173-180. Gomaa O.M. & Momtaz O.A. (2006). 16S rRNA characterization of Bacillus isolate and its tolerance profile after subsequent subculturing. Arab J. Biotech. 10(1): 107-116. Hamid R., Khan, M.A., Ahmad M., Ahmad M.M., Abdin M.Z., Musarrat J. & Javed S. (2012). Chitinases: An update. J. Pharm. Bioallied. Sci. 5(1): 21-29. Hoster F., Schimitz J.E. & Daniel R. (2005). Enrichment of chitinolytic microorganisms: isolation and characterization of a chitinase exhibiting antifungal activity against phytopathogenci fungi from a novel Streptomyces strain. Appl. Microbiol. Biotechnol. 66: 434-442. Holt J.G., Krieg N.R., Sneath P.H.A., Staley J.T. & Williams S.T. (1994). Bergey’s Mannual of Determinative Bacteriology 9. Lippincott Williams & Wilkins. pp. 559-569. Jabeen F. & Qazi J.I. (2014). Isolation of chitinase yielding Bacillus cereus JF68 from soil employing an edible crab shell chitin. J. Sci. Ind. Res. 73: 771-776. Kamil Z., Rizk M., Saleh M. & Moustafa S. (2007). Isolation and Identification of rhizosphere soil chitinolytic bacteria and their potential in antifungal biocontrol. GJMS. 2(2): 57-66. Kishore G.K., Pande S. & Podile A.R. (2005a). Biological control of late leaf spot of peanut (Arachis hypogaea) with chitinolytic bacteria. Phytopathology. 95: 1157-1165. Kishore G.K., Pande S. & Podile A.R. (2005b). Biological control of collar rot disease with broad- spectrum antifungal bacteria associated with groundnut. Can. J. Microbiol. 51: 123-132. Karunya S.K, Reetha D, Sarnaraj P. & Milton D.J. (2011). Optimization and purification of chitinase produced by Bacillus subtilis and its antifungal activity against plant pathogens. Int. J. Pharma. Biol. Archiv. 2: 1680-1685. Lim J.H, Rhee H.K, Kim Y.J., Lim K.B & vanTuyl J.M. (2003). Resistance to Fusarium oxysporum f.sp.lilii in Lilium. Acta Hortic. 620: 311-318. Lee J.Y., Moon S.S. & Huang B.K. (2003). Isoaltion and antifungal and antioomycete activities of aerugine produced by Pseudomonas fluorescens strain MM- B16. Appl. Environ. Microbiol. 69: 2023-2031. Li S., Zhang N., Zhang Z., Lou J., Shen B., Zhang R. & Shen Q. (2013). Antagonist Bacillus subtilis HJ5 controls Verticillium wilt of cotton by root colonization and biofilm formation. Biol. Fertil. Soils. 49: 295-303. Manoharachary C. & Mukerji K.G. (2006). Microbial activity in the Rhizosphere. Soil Biology. Mukerji K.G, Manoharachary C. and Singh J. (Eds.). Springer-Verlag Berlin Heidelberg. pp. 1-7. Mishra V.K. & Kumar A. (2012). Plant growth promoting and phytostimulatory potential of Bacillus subtilis and Bacillus amyloliquefaciens. J. Agric. Biol. Sci. 7(7): 509-519. Nam M.H., Park M.S., Kim H.G. & Yoo S.J. (2009). Biological control of strawberry Fusarium wilt caused by Fusarium oxysporum f.sp.fragariae using Bacillus velezensis BS87 and RK1 formulation. J. Microbiol. Biotechnol. 19: 520-524. Neeraja C., Anil K., Purushotham P., Suma K., Sarma PVSRN., Moerschbacher B.M. & Podile A.R. (2010). Biotechnological approaches to develop bacterial chitnases as bioshield against fungal diseases of plants. Crit. Rev. Biotechnol. 30(3): 231-241. Nguyen T.P., Nguyen T.P.O & Nguyen H.H. (2013). Antagonistic effect against Pyricularia oryzae of chitinase-producing bacteria isolated from rice rhizosphere. Cantho Uni. J. Sci. 31: 7-11. Nunez de Caceres Gonzalez F.F., Davey M.R., Sanchez E.C. & Wilson Z.A. (2015). Conferred resistance to Botry cinerea in Lilium by overexpression of the RCH10 chitinase gene. Plant Cell. Rep. pp. 1-9. Podile A.R. & Prakash A.P. (1996). Lysis and biological control of Aspergillus niger by Bacillus subtilis AF1. Can. J. Microbiol. 42:533-538. Sadfi N., Cherif M., Fliss I., Boudabbous A. & Antoun H. (2001). Evaluation of bacterial isolates from salty soils and Bacillus thuringiensis strains for the biocontrol of Fusarium dry rot of potato tubers. J. Plant Pathol. 83(2): 101-118. Shanmugaiah N., Mathivanan N., Balasubramania N. & Manoharan P.T. (2008). Optimization of Cultural contitions for production of chitinase by Bacillus laterosporous MMI2270 isolated from rice rhizosphere soil. Afr. J. Biotechnol. 7(15): 2562-2568. Setia I.N. & Suharjono. (2015). Chitinolytic assay and Identification of bacteria isolated from shrimp waste based on 16S rDNA sequences. Adv. Microbiol. 5: 541-548. Straathof T.P. & Van-Tuyl J.M. (1994). Genetic variation in resistance to Fusarium oxysporum f. sp. Lilii in the genus Lilium. Ann. Appl. Biol. 125: 61-72. Trachuk A.A., Revina L.P., Shemyakina T.M., Chestukhina G.G. & Stepanov V.M. (1996). Chitinases of Bacillus licheniformis B-6839: isolation and properties. Can. J. Microbiol. 42: 307-315. Veliz E.A., Martinez-Hidalgo P. & Hirsch A.M. (2017). Chitinase-producing bacteria and their role in biocontrol. AIMS Microbiol. 3(3): 689-705.