Plant growth promoting endophyte Burkholderia contaminans NZ antagonizes phytopathogen Macrophomina phaseolina through melanin synthesis and pyrrolnitrin inhibition

The endophytic bacterium Burkholderia contaminans NZ was isolated from jute, which is an important fiber-producing plant. This bacterium exhibits significant growth promotion activity in in vivo pot experiments, and like other plant growth-promoting (PGP) bacteria fixes nitrogen, produces indole acetic acid (IAA), siderophore, and 1-aminocyclopropane-1-carboxylate (ACC) deaminase activity. B. contaminans NZ is considered to exert a promising growth inhibitory effect on Macrophomina phaseolina, a phytopathogen responsible for infecting hundreds of crops worldwide. This study aimed to identify the possibility of B. contaminans NZ as a safe biocontrol agent and assess its effectiveness in suppressing phytopathogenic fungi, especially M. phaseolina. Co-culture of M. phaseolina with B. contaminans NZ on both solid and liquid media revealed appreciable growth suppression of M. phaseolina and its chromogenic aberration in liquid culture. Genome mining of B. contaminans NZ using NaPDoS and antiSMASH revealed gene clusters that displayed 100% similarity for cytotoxic and antifungal substances, such as pyrrolnitrin. GC-MS analysis of B. contaminans NZ culture extracts revealed various bioactive compounds, including catechol; 9,10-dihydro-12’-hydroxy-2’-methyl-5’-(phenylmethyl)- ergotaman 3’,6’,18-trione; 2,3-dihydro-3,5- dihydroxy-6-methyl-4H-pyran-4-one; 1-(1,6-Dioxooctadecyl)- pyrrolidine; 9-Octadecenamide; and 2- methoxy- phenol. These compounds reportedly exhibit tyrosinase inhibitory, antifungal, and antibiotic activities. Using a more targeted approach, an RP-HPLC purified fraction was analyzed by LC-MS, confirming the existence of pyrrolnitrin in the B. contaminans NZ extract. Secondary metabolites, such as catechol and ergotaman, have been predicted to inhibit melanin synthesis in M. phaseolina. Thus, B. contaminans NZ appears to inhibit phytopathogens by apparently impairing melanin synthesis and other potential biochemical pathways, exhibiting considerable fungistatic activity.

Introduction the environmental release of Burkholderia species as a biocontrol agent in agriculture is severely debated due to difficulties in distinguishing plant growth-promoting and pathogenic bacteria [20]. Recent developments in Burkholderia taxonomy and molecular analysis may help in answering questions concerning differences between the pathogenic and ecological properties of Burkholderia species in an attempt to reconsider the possibility of using selected strains for biocontrol [21].
The present study aimed to determine the safety of B. contaminans NZ as a biocontrol agent and its effectiveness against phytopathogens, especially M. phaseolina. The entire genome sequence was analyzed in this study to unravel information regarding the safety of B. contaminans NZ as a PGPR and biocontrol agent based on their siderophores and secondary metabolites producing capacities, and the absence of virulence genes. In vitro data also revealed the attributes of growth promotion. Moreover, Gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-mass spectrometry (LC-MS) analyses identified several potent antimicrobial compounds and antagonists of melanin synthesis, and the electron transport system produced by the bacterium in the presence of M. phaseolina. These compounds possibly antagonize fungal growth through different modes of action.

Bacterial, fungal strains and plant materials
The phytopathogenic fungi M. phaseolina was obtained from the Bangladesh Jute Research Institute (BJRI), Dhaka. Nigrospora sphaerica, Xylaria spp., Aspergillus fumigatus, Aspergillus niger, and Penicillium oxalicum were collected from the Molecular Biology Lab, Department of Biochemistry and Molecular Biology, University of Dhaka. Rhizoctonia solani was obtained from the Bangladesh Agricultural University, Mymensingh, Bangladesh. All the fungi were grown and maintained on potato dextrose agar (PDA) and for GC-MS analysis, M. phaseolina was grown in potato dextrose broth (PDB) (HiMedia, India) at 28˚C. The antagonistic bacterial strain was isolated from jute seed as an endophytic bacterium as described by Coombs and Franco [22,23] and identified as B. contaminans by 16S rRNA. Bacterial subculture was maintained on Tryptone Soy Agar (TSA) (HiMedia, India).
Jute seedlings were used to assess plant growth promotion activity of B. contaminans NZ. Fresh seeds of a jute variety (Corchorus olitorius var. O-4) were collected from BJRI. All tests were performed in triplicate if not mentioned otherwise.

Whole genome sequencing of B. contaminans NZ
B. contaminans NZ was cultured overnight in Tryptic Soy Broth (TSB) medium and incubated in an incubator shaker at 37˚C. The genomic DNA was extracted using the GenElute™ Bacterial Genomic DNA Kit (Sigma, Germany). A genomic library was constructed and employed for 300-bp paired-end whole-genome sequencing at the Genome Research Institute of North South University, Dhaka, Bangladesh using an IlluminaMiSeq platform (Illumina, San Diego, CA, USA) according to the manufacturer's instructions. The generated raw reads (10xcoverage) were assembled using SPAdes version 3.11 [24]. The generated scaffolds were mapped and ordered using ABACAS-a reference-based assembler [25], considering Burkholderia contaminans CH1 as the reference genome [GenBank accession no. AP018357 to AP018360 (four entries)] [26].
Genome sequence analysis and annotation. For structural annotation of the genome Rapid Annotations using Subsystems Technology (RAST) server [27] was used (http://rast. nmpdr.org) along with SEED database for the functional annotation of predicted gene models. SEED also provides subsystem (collection of functionally related protein families) and derived FIGfams (protein families) which represent the core of RAST annotation engine [28]. Subsequently, genome annotation through PIFAR annotation tool was used for the identification of bacterial genetic factors involved in plant host interaction [29]. Moreover, antiSMASH bacterial version 5.0 (https://antismash.secondarymetabolites.org/#!/start) [30] analysis was carried out to identify the biosynthetic gene clusters for secondary metabolites of B. contaminans NZ.
Examining the presence of proteins fostering virulence and pathogenicity. A total of 147 Burkholderia specific proteins reported for pathogenicity and virulence were collected from Virulence Factor Database (VFDB) (http://www.mgc.ac.cn/VFs/main.htm) [31] and investigated for the presence of virulence and pathogenicity related proteins in B. contaminans NZ with the help of BLASTp considering the e-value of at least 1e-10 and sequence identity, not less than 70%.
Data deposition. This Whole Genome Shotgun project has been deposited in DDBJ/ ENA/GenBank under the accession QRBC00000000. The version described in this paper is version QRBC01000000.

In vivo growth promotion study of B. contaminans NZ on jute
Inoculation of jute seeds with the endophyte. At first B. contaminans NZ was cultured on TSA plates and then inoculated in TSB media and incubated at 37˚C at a shaking speed of 180 rpm for 24-48 h. At the same time, scarification was done for jute seeds with sandpaper and the seeds surface was sterilized using 5% NaOCl. Sterilized seeds were dipped into B. contaminans NZ suspension. From McFarland standard, bacterial cells of 10 8 CFU/ml were determined by a serial dilution method [32].
In vivo pot experiment in hydroponic culture. Jute seeds inoculated with a bacterial suspension were grown under controlled environmental conditions in a plant growth chamber (Weiss Technik India Private Limited) at 28˚C, 70% relative humidity, and 16-h light/8-h dark cycle. The seeds were allowed to germinate for three days and later grown in a hydroponic culture system.  50.0g per litre) [33] was added to the seedlings. pH of the solution was adjusted to 5.5 with NaOH. Thirty seeds were sown 1 cm deep in a cork sheet in separate pots. Jute plants were grown for 10 days in hydroponic culture and 5 plants were collected for the measurement of fresh and dry weights, shoot, and root lengths on 4, 7, and 10 days after transfer of the seedlings into the hydroponic solution. The experiment was repeated three times.

Antifungal activity assay
Dual culture assay. In vitro bacteria-fungal dual-culture assays were established by both agar well diffusion and cross streak methods in 9 cm diameter Petri dish systems containing PDA medium [34]. A 5 mm plug taken from the plate of an actively growing fungal colony was inoculated on one side of the Petri dish. Fresh cells of B. contaminans NZ were either streaked in 3 cm length parallel lines on the other side of the fungal plug or 20 μl of the overnight bacterial liquid culture was introduced into a 6 mm diameter well punched aseptically with a pipette tip in an agar plate. Control treatments containing only the fungus were also set up. The plates were incubated at 28˚C for 4 to 5 days. The experiment was set up for four replicates. After incubation, the diameters of the fungal colonies were scored and measured.
Antifungal activity of B. contaminans NZ was determined against all the studied phytopathogenic fungi and the toxicity was expressed as a percentage of growth inhibition (PGI) and calculated according to the Zygadlo et al. [35] formula, where GC represents the average diameter of the fungus grown on PDA (control); GT represents the average diameter of the fungus co-cultivated on the PDA dish with the antagonistic bacterium. A paired t-test was used to check whether the percentage of growth inhibition was significant.
Co-culture of M. phaseolina and B. contaminans NZ. For GC-MS analysis, M. phaseolina was cultured in 20 ml and 100 ml PDB media in two separate conical flasks. B. contaminans NZ was cultured in 80 ml and 100 ml of TSB media in two separate conical flasks.
On the second day of incubation, 20 ml M. phaseolina culture was mixed with 80 ml B. contaminans NZ to form the co-culture and put to incubation again. All the cultures were incubated at 28˚C for 5 days for volatile compound extraction and observed for any phenotypic changes in M. phaseolina culture in the presence of B. contaminans NZ.
For HPLC and LC-MS analysis, a separate 1000 ml culture of B. contaminans NZ was cultured under similar conditions described above.

Extraction of organic compounds from culture media
On day 6 of single cultures of fungus, bacterium, and co-cultured media were filtered using Whatman filter paper and centrifuged at 4000 rpm for 7 minutes to remove cell debris. Supernatants were separated into individual conical flasks. 1:1 ethyl acetate (v/v) was added to each flask, shaken, and kept for 4 hours. The organic phases were separated and dried using a rotary evaporator. Finally, the extracts were dissolved in 2 ml of ethyl acetate. A separate extraction (for HPLC and LC-MS analysis) of B. contaminans NZ culture media was also conducted using n-hexane instead of ethyl acetate and shaken for 24 hrs. The extracted fraction was finally dissolved in methanol.
GC-MS analysis of ethyl acetate extract. Ethyl acetate extract of extracellular components was subjected to GC-MS analysis using Shimadzu GCMS-QP2010 Ultra; (Japan) mass detector connected with a capillary column of Rxi-5ms, 30 m long, 0.25 mm i.d.0.25 μm film thickness. 1μL of the sample was injected with splitless mode. Helium was used as the carrier gas with the flow rate was set at 1.0 ml/min. The column oven temperature was programmed from 40˚C (1 min hold) at a rate of 10˚C/min to 300˚C (10 min hold). Injector temperature was maintained at 250˚C. Temperatures of the mass interface and ion source were 250˚C and 200˚C respectively, and detector voltage was 0.4 KV. The analysis was carried out in the EI (electron impact) mode with 70e V of ionization energy. The analysis was performed in full scan mode, ranging from m/z 50 to 400. The compounds were detected after analyzing the mass spectrum of each component using the NIST11 library.
HPLC and LC-MS analysis of methanolic extract. Analytical HPLC Dionex Ultimate 3000 with a C18 column (Nucleodur, 250x4.6 mm, particle size 5μm, pore size 110Å, 100-5C18ec) was used to analyze the bacterial extract dissolved in methanol. After dissolution, the extract was filtered and 200 μL was injected into the column. A multi-step gradient system (solvent system: acetonitrile and water with 0.05% trifluoroacetic acid; flow rate: 1ml/min; temperature: 23˚C) was performed with an optimized protocol (0-10 min, 10% acetonitrile; 10-30 min, 10-80% acetonitrile; 30-40 min, 80% acetonitrile; 40-50 min, 80-0% acetonitrile; volume fraction). The wavelength at 225 nm was used to detect compounds eluted from the column. Various fractions were collected containing high peak intensity in the chromatogram and the fractions were dried at low temperature in a lyophilizer (LYOVAC GT 2) and dissolved in methanol to test for antifungal activity and followed by a re-run in HPLC to confirm the reproducibility of the peak. The purified active fraction was then subjected to LC-MS.
For Liquid Chromatography (LC), Agilent 1290 Infinity II instrument was used. C18 column (2.1 mm x 100 mm x 1.8 um) (Zorbax RRHD Eclipse) was used as the stationary phase and 50:50 ratio of water (with 0.1% formic acid) and acetonitrile was used as the mobile phase on isocratic mode. The column temperature and the mobile phase flow rate were kept at 30˚C and 0.2 mL/min throughout the program. The sample injection volume was 5 uL. For Mass Spectrometry (MS) analysis, an Agilent 6420 Triple Quad Mass Spectrometer System was used, equipped with an electrospray ionization source operating at positive mode and scanning from m/z 100 to 1000. Nitrogen was used as the nebulizer gas, with pressure and flow set at 45 psi and 11 L/min, respectively. The capillary voltage was maintained at 4000 V, and dry gas temperature at 350˚C. Mass Hunter software was used to control and analyze the data. The compound was identified based on its MS fragmentation and analyzed according to literature [36].

Identification of homologous melanin pathway(s) in M. phaseolina
Fungi contain different pathways for melanin biosynthesis; among them two pathways namely 1,8-dihydroxynaphthalene (DHN)-melanin and L-3,4-dihyroxyphenylalanine (DOPA)-melanin pathways, are primarily found in the ascomycota group of fungi [37]. M. phaseolina proteins homologous to this pathway were searched using BLAST. Pathway proteins discovered through literature review and BLAST searches were tested against the M. phaseolina proteome. At least a 30% identity cut-off was taken to consider the homology between the target and query protein [38].

Statistical analysis
The average shoot length, average root length and average height from the three repeated experiment of in vivo pot experiment was taken for statistical analysis. Data obtained were subjected to analysis of variance (ANOVA) test against the control. The results presented as average means, standard deviation (SD), and standard error (SE) were determined by following the standard procedures.

Pot tests for in vivo assay of plant growth promotion
B. contaminans NZ significantly increased the root and shoot lengths, and the average height of jute seedlings (P<0.05) (Fig 1, S1 and S2 Tables). The maximum root length of 11.28 cm, was 116.9% greater than that of the control.
The effect on shoot length was also found to be significant; on average, it was 50.7% longer than that of the control jute plants. Fresh weight of 4-, 7-, and 10-day old jute seedlings increased by 141%, 122%, and 41%, respectively, compared to the controls (P<0.05).
In the presence of B. contaminans NZ, the dry weight of jute plants on an average, increased by 49.8% in all three replicates for each of the 4-, 7-, and 10-day measurements.

Whole genome data analysis
A summary of the whole genome annotation of B. contaminans NZ sequenced using Illumina MiSeq technology is presented in S3 Table. Functional analysis of B. contaminans NZ using the RAST server (Rapid Annotation using Subsystem Technology, http://rast.nmpdr.org), antiSMASH, and PIFAR (plant-bacteria information factor resource) revealed numerous genes and gene clusters ( Table 1, S4 Table), which are reported to be exclusively associated with plant growth promotion.
As observed in other PGPR Burkholderia strains [39], the nifHDK operon required for nitrogen fixation was detected in B. contaminans NZ, along with the 1-aminocyclopropane-1-carboxylate (ACC) deaminase coding sequence, gene coding for iron(III) ABC transporter substrate-binding protein, phytoene synthase gene, the coding sequence for phosphotransferase system (PTS), and major facilitator superfamily (MFS) transporter genes. In silico analysis using RAST also revealed the presence of genes for indoleacetamide hydrolase involved in IAA  production, which is a plant hormone associated with plant growth [40], and the eno gene, which assists in phosphate solubilization. Genome analysis also indicated the presence of pyrroloquinoline quinone synthase and glucose dehydrogenase, which are implicated in gluconic acid and 2-ketogluconic acid production, and required for mineral phosphate solubilization [41]. The annotated genome of B. contaminans NZ also contains 30 siderophore-related genes associated with ferric siderophore transport, ABC-type siderophore export system, arthrobactin, siderophore pyoverdine, and an intact siderophore pyochelin biosynthesis pch gene cluster. From the PIFAR data, in B. contaminans NZ, the biosynthesis of the major siderophore pyochelin is apparently associated with five genes (PMID: 22261733) distributed over five different regions in the bacterial genome. While examining the core biosynthesis genes, genetic loci related to antibiotic production were also found in B. contaminans NZ, as presented in Table 1. Moreover, genes for pyrrolnitrin, an antifungal secondary metabolite produced by certain Bcc and several other gram-negative bacteria, were also identified. This metabolite inhibits the growth of a wide range of fungi and pathogenic bacteria [42,43].

Most virulence genes absent in B. contaminans NZ
Various virulence factors reported in pathogenic bacteria and other plant endophytic strains of Burkholderia (B. contaminans CH1, B. contaminans MS14) were screened for in the B. contaminans NZ genome, which failed in identifying major virulence-related genes, such as the biosynthesis genes for cable pili, toxoflavin, cepacian, and O-antigen (O-Ag) biosynthetic cluster ( Table 2). This differs from pathogenic strains that contain many virulence genes. However, few virulent genes in B. contaminans NZ are present as incomplete gene clusters.

Antifungal activity assay
In dual culture, B. contaminans NZ substantially inhibited the growth of six plant fungal pathogens, namely Nigrospora sphaerica, Xylaria spp., Aspergillus fumigatus, Aspergillus niger, Penicillium oxalicum, and Rhizoctonia solani (Fig 2). In dual-culture assay, B. contaminans NZ demonstrated significant growth inhibitory activity, which ranged from 44% to 58% than that of the control (P<0.05), when co-cultured with all the tested fungi. The growth inhibition rates observed for each fungal species are shown in Table 3.

Growth suppression and chromogenic aberration in B. contaminans NZ challenged M. phaseolina
M. phaseolina inoculated in liquid culture was initially pale in color, which assumes a characteristic black color within two to three days that perpetuates until day 5. However, a

PLOS ONE
M. phaseolina culture challenged by B. contaminans NZ deviates from this usual behavior and retains the initial pale color up to day 5. Furthermore, M. phaseolina growth when subjected to co-culture with the bacterium remained practically static from day 2. Fig 3 shows the phenotypic changes in fungal growth from days 2 to 5.

Compound Name
Culture Name

Retention Time (min)
Biological Activity Reference (s)

Identification of putative melanin pathways in M. phaseolina through homology-based search
After an extensive literature review of different melanin pathways active in fungi, polyketide synthase, tetrahydroxy naphthalene reductase, trihydroxy naphthalene reductase, scytalone

Discussion
Since chemical control and most cultural practices are not effective representative tools that limit the pathogenic fungal growth and their distribution [69], the use of biocontrol agents has become a promising alternative in agriculture. Biocontrol using plant growth-promoting rhizobacteria (PGPR) is a potentially attractive and efficient disease management approach, as it

PLOS ONE
promotes plant growth, enhances tolerance to abiotic stress [70], suppresses pathogens locally, and induces systemic resistance (ISR) against a broad range of crop diseases [71,72]. Several studies conducted on endophytes have emphasized their ability in promoting plant growth and their additive/synergistic effects on plant growth and protection. In general, such growth-promoting rhizospheric bacteria belong to the following species: Alcaligenes, Arthrobacteria, Azospirillum, Azotobacter, Bacillus, Burkholderia, Enterobacter, Klebsiella, and Pseudomonas [8,73]. The endophytic strain B. contaminans NZ isolated from jute has also been characterized in vivo for its potential plant growth-promoting (PGP) traits, which significantly increased both yield and biomass when jute seedlings were inoculated with the bacterium, and notably reduced the development of disease symptoms caused by M. phaseolina.
As B. contaminans NZ exhibited growth promotion and apparent protection against the phytopathogen M. phaseolina, its whole genome was sequenced to determine the characteristic genetic features. Genome annotation of B. contaminans NZ assisted in identifying genes related to phytohormone production, stress alleviation, phosphate solubilization, nitrogen fixation, antibiotic biosynthesis, and siderophore biosynthesis. In vitro experiments of the endophyte also displayed significant production of the above-listed compounds (S5 Table, S2 Fig).
Certain Burkholderia species demonstrate clinical associations and plant-pathogenic traits [74,75]; therefore, genes related to virulence and pathogenicity were also investigated in the B. contaminans NZ genome. Several bacterial species from the B. cepacia complex (Bcc) are opportunistic pathogens [76]. B. cenocepacia strains expressing both cable (Cbl) pili and 22-KDa adhesin proteins have been reported to bind strongly to cytokeratin 13 (CK13), which efficiently infects the host cell by invading the squamous epithelium [77]. However, the cblA (giant cable pili) gene was absent in the symbiotic and legume-nodulating species [78]. B. contaminans NZ also lacks the cable pili biosynthesis gene cluster, indicating their inability to attach to the host cell, which is required to initiate infection. Furthermore, VgrG-5, which is a Burkholderia type VI secretion system 5-associated protein, required for complete mammalian virulence [79], is also absent in B. contaminans NZ genome. Virulence-associated protein sequences (for chemotaxis, attachment, type 3 and type 6 secretion systems; T3SS and T6SS) and typically described hallmark features representing the true and opportunistic pathogenic Burkholderia strains are also absent in B. contaminans NZ. Cystic fibrosis (CF)-related O-antigen of lipopolysaccharides associated with transmitting infections in CF patients [80] and zinc metalloproteases that may be involved in the overall virulence of several Bcc strains [81] are also absent. Among the virulence-associated proteins of Burkholderia, only the sodC gene encoding a Cu 2+ and Zn 2+ containing periplasmic superoxide dismutase that contributes in intracellular survival, which indicates self-protection ability in CF patients, is present in the genome [82]. Burkholderia species commonly produce plant-toxic secondary metabolites, polysaccharides, and other toxins such as rice grain rot and wilt causal agent, toxoflavin; exo-
Growth inhibition of M. phaseolina by B. contaminans NZ caused marked morphological changes and chromogenic aberration in the phytofungus [13]. Such morphological changes in the cell membrane have also been reported for B. cepacia-mediated inhibition of F. solani and C. dematium [83]. Burkholderia CF66I noticeably alters R. solani hyphae with multiple branches and swelling [84]. In previous reports, co-culturing M. phaseolina and B. contaminans NZ in solid media caused the fungus to lose its characteristic black color, restricted its growth, and attenuated its infectivity [13]. The present study also observed a similar deviation in pigmentation and growth repression in liquid co-culture (Fig 3).
Burkholderia spp. reportedly produces potent antifungal compound pyrrolnitrin [89]. This compound exhibits inhibitory effects by obstructing the synthesis of vital biomolecules (DNA, RNA, and protein), uncoupling oxidative phosphorylation, impeding mitotic division, and inhibiting several biological mechanisms. The genome of B. contaminans NZ also contained a gene cluster for pyrrolnitrin synthesis ( Table 2). Although the GC-MS data revealed the presence of several compounds that implied possible modes for M. phaseolina growth suppression, pyrrolnitrin in the secondary metabolite profile was absent. Therefore, the extraction process was altered by substituting the extraction solvent with n-hexane instead of ethyl acetate to identify pyrrolnitrin, which is in agreement with earlier reports that employed a similar method [43]. Thereafter rest of the method remained similar except that the extract was dissolved in methanol instead of ethyl acetate. RP-HPLC was performed to purify the active compound(s) to detect the appreciable suppression of fungal growth by the crude extract. Only one fraction exhibited considerable inhibitory activity against M. phaseolina, and its mass was identified using LC-MS analysis (257.1 Da), which was identical to the standard pyrrolnitrin (molecular weight: 257.07 g/mol).
Further investigation of the M. phaseolina enzymes involved in melanin synthesis, similar to those reported earlier in other fungi, led to the identification of SDR-short chain dehydrogenase/reductase, which is similar to both trihydroxy naphthalene reductase [90] and tetrahydroxy naphthalene reductase [91], which are enzymes of the DHN-melanin pathway [92]. Tyrosinase [92,93] performs multiple catalysis in the DOPA-melanin pathway; it converts phenylalanine to tyrosine, tyrosine to DOPA, and DOPA to DOPA quinone, which is eventually converted to melanin further downstream. However, both DOPA-melanin and DHN-melanin pathways are yet to be fully elucidated. This reduces the scope of finding other homologous pathway proteins in M. phaseolina. The percent identity of beta-ketoacyl synthase and tyrosinase was lower (~31%) than that of tetrahydroxy naphthalene reductase, trihydroxy naphthalene reductase, and scytalone dehydratase (~75% or more). This was possibly due to only few reports that describe the two genes with precise genomic features. Based on this evidence, a tentative melanin pathway was proposed for M. phaseolina in this study (S1 Fig).

Conclusion
Whole genome analysis and secondary metabolite characterization have acknowledged B. contaminans NZ as a good biocontrol agent, in addition to its role in plant growth promotion. It suppresses the growth of multiple fungi, especially the phytopathogen M. phaseolina, by using catechol, pyrrolnitrin, and other antimicrobial agents. Compounds identified in the extracts of B. contaminans NZ or the co-culture of B. contaminans NZ and M. phaseolina contain compounds that inhibit melanin biosynthesis, which possibly contributes to the observed chromogenic aberration and growth suppression of the fungi. B. contaminans NZ can be established as a bio-control agent by conducting toxicity tests to ensure its safety, followed by field experiments to determine its efficacy under different environmental conditions. Hence, further studies are needed to optimize the formulation and application methods of B. contaminans NZ to fully maximize its potential as an effective agent in controlling M. phaseolina.