Secondary Metabolism Gene Diversity and Cocultivation toward Isolation and Identification of Potent Bioactive Compounds Producing Bacterial Strains from Thailand's Natural Resources

Thailand was proposed to be rich unexplored source of microorganisms, especially bacterial strains. There should be bacteria with high secondary metabolite production potential in the natural resources that are still unidentified. Moreover, they might not produce secondary metabolites in standard laboratory culture condition after isolation, in which coculture condition would help us pursuing the bacteria to produce bioactive metabolites. Here, we aimed to identify new bacterial strains with high secondary metabolite production potential from Thailand's natural resources. To achieve the goal, we performed bacteria isolation, phylogenetic analysis, degenerate PCR of secondary metabolism genes, cocultivation, antibacterial analysis, and HPLC chemical profiling. We isolated distinct 40 bacterial strains, which have over 98% 16S rRNA sequence similarity with known species. There were 22, 31, and 29 strains giving positive PCR amplification of NRPS, PKS, and TPS genes, respectively. Among them, Bacillus licheniformis RSUCC0101 had the highest number of PCR products, 26. In standard single culture condition, crude extracts prepared from Bacillus safensis RSUCC0021 and Bacillus amyloliquefaciens RSUCC0282 could inhibit the growth of Staphylococcus aureus ATCC25923. Furthermore, the cocultivation and HPLC analyses showed that the extracts prepared from 3 pairs of culture between Staphylococcus sp. RSUCC0020, Micrococcus luteus RSUCC0053, Staphylococcus sp. RSUCC0087, and Staphylococcus pasteuri RSUCC0090 could inhibit the growth of Staphylococcus aureus ATCC25923 and produced distinct chemical profiles from their single culture condition. Our study led to the isolation and identification of several promising bacterial strains for production of secondary metabolites that might be useful in biomedical applications.


Introduction
ere are several problems around the world including suspension of petroleum energy source, unresolved pollution, climate change, emerging diseases, untreated diseases, and drug resistance pathogens. We direct our expectation toward the natural resources, which might possess unexplored natural compounds, to combat the mentioned problems [1][2][3][4]. Considering the natural compounds, the diverse chemical structures of secondary metabolites produced by microorganisms were proposed to be the valuable metabolites that act against the harmful biochemicals or proteins. Among the microbes, the easily evolving bacteria especially possess the diverse ability to produce the diverse secondary metabolites through their biosynthesis pathways. Genome-wide studies revealed that the bacteria species could synthesize groups of secondary metabolites including nonribosomal peptides (NRP), polyketides (PK), terpene (TP), and hybrid PK/NRP [5][6][7]. e bacteria use secondary metabolite as biological weapon against the other species in the same niche; protect themselves from physical environment, communication substances, and symbiosis stimulant, or even are hormones or pheromones to other organisms. Accordingly, novel secondary metabolites identified from the unexplored bacterial strains might be used as beneficial compounds for both industry and medical applications [8,9].
In order to isolate a new bacterial strain from natural specimen of interest, we need appropriate protocols for isolation, identification, and maintenance of the bacterial strains from the natural resources. ese include the use of appropriate isolation media depending on the niche of the bacteria target to promote and accelerate the growth of the bacterial strain but not from the standard media. Afterward, standard PCR is used for amplifying 16S ribosomal RNA gene fragment from bacterial isolated genomic DNA. Subsequently, the PCR product is then sequenced and compared with the 16S ribosomal RNA gene of the closely related bacterial strains using phylogenetic analysis [10]. By using degenerate primer pairs designed for the specific amplification of each class of the genes related to the biosynthesis of secondary metabolites, the number of gene fragments reflects the potential of the bacteria to synthesize the secondary metabolites for their own benefits [11]. Consequently, the new identified species with high number of genes that might be involved in the biosynthesis of secondary metabolites exhibit more chance to produce unidentified secondary metabolites with novel functions or bioactivities. Not only the degenerate PCR amplification but also the genome sequence data mining became the new era for the discovery of new genes that might be involved in the biosynthesis of new secondary metabolite from bacterial species. ere are enormous numbers of genes identified from the genome sequence data up to date; however, only few of them were characterized for their biosynthetic functions [12][13][14]. e new identified species with high secondary metabolite production potential genes might not produce any bioactive compound in the standard laboratory condition, that is, the normal situation found in several studies. is is because the genes that might be involved in secondary metabolism are not expressed in the standard culture condition. erefore, culture condition optimization or genetic engineering processes are needed for activation of the biosynthesis pathway [15]. A study in 2017 showed that a simple cocultivation technique between new identified bacteria with a fungus could stimulate the production of surfactant from the unproduced single culture condition. e explanation behind the situation would be the physiological response of the bacteria to the competitor in the same niche. e cocultivation technique should be also considered for the analysis of bioactivity of the newly identified bacterial strains [16].
ere are only 15% of bacterial strains in the world that were isolated, identified, and used in biotechnological laboratory. In particular, ailand, which is located in an appropriate residential zone for bacterial strains, might be rich in unexplored natural bacterial resources such as the sea, soil, and forest. Accordingly, unidentified secondary metabolites from the bacterial species should be explored as well [17]. Accordingly, in this work, we aim to identify new bacterial strains with high secondary metabolite production potential from our natural resources including seawater, soil, forest wood, and herbs. Furthermore, we also aim to identify their ability to produce compounds that might be useful in medical application using both single standard culture condition and cocultivation culture condition. We expect that the new identified species with high secondary metabolite production potential would be a promising bacterial strain that could produce biomedical valuable secondary metabolite.

Isolation and Maintenance of Bacterial Strains.
e natural specimens consisting of 100 mg of Bangkhuntien's mangrove forest soil, Nam Nao National Park's soil, and Yaowarat Chinese market herbs were mixed with sterile distilled water to final volume of 10 mL. For Rayong's seawater, 10 mL was collected and diluted 1,000 times with NSS before use. After homogenization, 100 µL of the aqueous solutions was plated on agar plate of CDA, MEA, MRS, NA, PCA, PDA, SDA, and TSA (Difco, USA). We incubated the plate 3 days and then transferred single colony into new NA agar plate. During doing bacteria culture, bacterial Gram of all strains was identified by standard Gram staining. In this study, we used 4 importance human pathogens as standards for antibacterial assay, namely, Bacillus cereus ATCC14579, Escherichia coli ATCC25922, Pseudomonas aeruginosa ATCC27853, and Staphylococcus aureus ATCC25923 (ATCC, USA). In order to maintain all the isolated and standard bacterial strains, we kept their 3day NB cultures with 25% glycerol and restocked them every 6 months at −80°C. All bacteria strains' genomic DNA was prepared using Presto ™ Mini gDNA Bacteria Kit following the manufacturer protocol (Geneaid, Taiwan).

Phylogenetic Analysis of Amplified 16S rRNA Gene
Fragments. We used 16S27F (5′-AGAGTTTGATCCTG GCTCAG-3′) and 16S1492R (5′-GGTTACCTTGTTAC-GACTT-3′) as primers for amplification of 16S rRNA genes from all isolated bacterial genomic DNA [18]. A PCR reaction consists of 10 ng genomic DNA, 1X PCR buffer, 2.5 mM MgCl 2 , 1 mM dNTP, and 1 unit of DNA polymerase (Biotechrabbit, Germany). We used the initial denaturation at 94°C for 2 min, followed by 35 rounds of 94°C for 30 sec, 54°C for 30 sec, and 72°C for 2 min before final extension at 72°C for 5 min as thermal cycler condition. e amplified DNA fragments were then purified using GenepHlow Gel/ PCR Kit (Geneaid, Taiwan), and subsequently their DNA sequence was determined via sequencing service of SolGent company (SolGent, Korea). e DNA sequence data were initially compared with NCBI database using BLAST. DNA of closely related species was then retrieved from the database; we used ClustalX for multiple sequence alignment and then reconstructed phylogenetic trees by programs in PHYLIP package. e unrooted phylogenetic trees were constructed using neighbor-joining method under Kimura 2-parameter model. Newick's standard files were visualized by FigTree v1.3.1. Bootstrap values obtained from 1,000 replicate pseudosamples were overlaid onto the visualized phylogenetic trees in percentages of the consensus.

Crude Extract Preparation from Single Cultivation and Cocultivation.
A single colony of each bacterial strain was used for inoculum preparation. A 5 mL NB overnight culture was then diluted with NSS until the turbidity reached McFarland's standard No. 0.5 (approximately 1.5 × 10 8 cells/ mL). One mL of the inoculum was then added to 100 mL of freshly prepared NB media and further incubated at 37°C for 7 days. For cocultivation, 2 of the inoculums were added to the same 100 mL NB media and further incubated at the same condition as the single cultivation. After that, the cultured media were then collected by centrifugation and extracted by equal volume of ethyl acetate in separation funnel, and the ethyl acetate fraction was then collected. Subsequently, we used rotary evaporator to evaporate the collected fractions into the crude extract. e crude extract of each bacterial strain was prepared into triplicates.

Disc Diffusion Assay. Bacillus cereus ATCC14579,
Escherichia coli ATCC25922, Pseudomonas aeruginosa ATCC27853, and Staphylococcus aureus ATCC25923 (ATCC, USA) were used as pathogenic representatives for antibacterial screening in this work. e NSS inoculum of each standard bacterium with approximately 1.5 × 10 8 cells/ mL was swabbed on NA. Five mg of each prepared crude extract was impregnated into sterile 6 mm diameter filter paper discs (Whatman, UK) before being placed onto the swabbed NB. e discs containing ampicillin (10 μg/disc), ceftriaxone (30 μg/disc), chloramphenicol (30 μg/disc), and rifampicin (5 μg/disc) (Oxoid, UK) were used as positive control for Escherichia coli, Pseudomonas aeruginosa, Bacillus cereus, and Staphylococcus aureus, respectively. After incubation for overnight, the antibacterial activity of each crude extract was determined by measuring the IZD in mm.

Diversity of Bacterial Strains Isolated from ailand's Natural Resources.
We could isolate 40 morphological distinct bacterial strains from ailand's 4 natural sources: 16 strains from Bangkhuntien's mangrove forest soil, 1 strain from Nam Nao National Park's soil, 20 strains from Rayong's seawater, and 3 strains from Yaowarat Chinese market herbs. After we sent and obtained the data of approximately 1,400 bp of the 40 strains' amplified 16S rRNA gene fragments, we could assign the putative species of all the strains using general basic local alignment search tool (BLAST) to search and prepare phylogenetic analysis. BLAST analysis showed that their 16S rRNA sequences had over 98% similarity with the sequence of known species stored in National Center for Biotechnology Information (NCBI) database. e phylogenetic analysis with high confidential value constructed from 1,000 pseudosamples also supported the correct identification of the species (Figure 1). ere are 8 strains that were identified as putative pathogenic bacteria: Bacillus cereus RSUCC0142, Bacillus subtilis RSUCC0029, Staphylococcus haemolyticus RSUCC0013, Staphylococcus pasteuri RSUCC0056, Staphylococcus pasteuri RSUCC0090, Vibrio fluvialis RSUCC0003, Vibrio fluvialis RSUCC0009, and Vibrio fluvialis RSUCC0010. e highest diverse genus isolated from the 4 resources in our study is Bacillus, which accounted for 37.5% of total isolated strains. We could isolate more Gram-positive bacteria (28 strains) than Gram-negative bacteria (12 strains) ( Table 1).  highest number of amplified secondary metabolism gene fragments in this study is Bacillus licheniformis RSUCC0101. We could amplify 26 PCR products from its genomic DNA using 3 primer pairs consisting of 6 NRPS A domains, 9 PKS KS domains, and 11 TPS cyclases (   Figure 2).

Discussion
Bacteria are the largest diverse microorganisms found in the world. ey have a major role as the digester in the ecosystem. ey transform macromolecules into small metabolites that can be absorbed easier by other organisms in the same niche. In order to maintain their relationship with the environment, bacteria might produce secondary metabolites as antibiotics, protectants, stimulants, hormones, and pheromones [19]. Because of the unexplored bacterial strains and their secondary metabolite diversity, the undiscovered bacteria in ailand's natural resources might be a promising source of new bioactive secondary metabolites [17]. Consequently, we initiated our project to explore, isolate, and identify high potential secondary metabolite producing bacterial strains from Bangkhuntien's mangrove forest soil, Nam Nao National Park's soil, Rayong's seawater, and Yaowarat Chinese market herbs. We successfully isolated and identified 40 bacterial strains from the mentioned ailand resources. Using general BLAST search and phylogenetic analysis, we assigned the genus and species names to all of the strains (Supplementary Material 6). We could identify more Gram-positive (28 strains) than Gram-negative (12 strains) bacteria from the natural resources similar to the study in 2014. e ratios of bacteria diversity of the current study were 70% and 30%, but 61% and 39% in the previous study, for Gram-positive and Gram-negative bacteria, respectively [20].
e Gram-positive bacteria usually play the major role in carbon, nitrogen, and 6 Scientifica phosphorus cycle. Accordingly, this might be the reason why we could isolate diverse Gram-positive bacterial strains from the sea, soil, and tree resources [21]. We could amplify A domain of NRPS from 22 bacterial strains and KS domain of PKS gene from 31 bacterial strains, respectively. Interestingly, half of them that give positive PCR amplification results belong to Bacillus. It is well known that Bacillus genus is defined as high potential NRP, PK, and hybrid PK/NRP producer in previous genome mining [22]. Especially for our strain, Bacillus licheniformis RSUCC0101, its genome should contain very high secondary metabolism gene paralogs for production of NRP (6 A domains), PK (9 KS domains), and TP (11 cyclases) revealed by our degenerate PCR amplifications. e result showed the potential of the strain to be a high potent secondary metabolite producer similar to previous bioinformatics genome mining study [23]. However, the amplification of KS domain from Staphylococcus species yielded negative PCR products except Staphylococcus sp. RSUCC0008. We hypothesized that the positive gene fragment amplification from Staphylococcus sp. RSUCC0008 could be a result of horizontal gene transfer from the other bacterial species within the same niche [24].
is hypothesis should be proven by further full-length PKS gene amplification, gene sequencing, genome sequencing, and phylogenetic analysis. For TPS gene amplification, we expected higher TPS gene diversity in Gram-negative than Gram-positive bacteria, which was discussed previously in 2015 [25]. However, our results showed no significant difference in TPS gene diversity between Gram-positive and Gram-negative bacteria. Furthermore, we could not amplify TPS gene fragments from Pseudoalteromonas byunsanensis RSUCC0073, Pseudoalteromonas piscicida RSUCC0088, Vibrio azureus RSUCC0093, and Marinomonas fungiae RSUCC0100.
Bacillus safensis was first isolated from Mars Odyssey Orbiter spacecraft. It was reported to synthesize several carotene terpenoid substances for protecting its cell from UV irradiation [26]. In our work, the extract prepared from single culture of Bacillus safensis RSUCC0021 could inhibit the growth of Staphylococcus aureus ATCC25923 with IZD value of 8.0 ± 1.30 mm.
is bioactivity should belong to NRP, PK, or hybrid PK/NRP rather than TP. Accordingly, we amplified 2 A domains of NRPS gene from the strains, which might be involved in the biosynthesis of the antibacterial compounds identified in 2014 [27]. e extract prepared from Bacillus amyloliquefaciens RSUCC0282 could also inhibit the growth of Staphylococcus aureus ATCC25923 with IZD value of 19.0 ± 1.50 mm. However, there is no previous report showing that Bacillus amyloliquefaciens could produce secondary metabolite compound against the growth of Staphylococcus aureus. Bacillus amyloliquefaciens was usually used as a biocontrol agent that can stimulate plant immune system against pathogenic fungi Ralstonia solanacearum, Rhizoctonia solani, and Alternaria tenuissima. Moreover, Bacillus amyloliquefaciens could produce barnase and NRP plantazolicin that specifically inhibit the growth of Bacillus anthracis [28]. e antibacterial activity of the extracts prepared in our study against Staphylococcus aureus would need further investigation for chemical structure elucidation of the bioactive compounds. Surprisingly, we could not observe bioactivity of any extracts prepared from coculture of Bacillus safensis RSUCC0021 and Bacillus amyloliquefaciens RSUCC0282 against Staphylococcus aureus ATCC25923. It is possible that the growth of Bacillus safensis RSUCC0021 and Bacillus amyloliquefaciens RSUCC0282 was inhibited and/or obtained inappropriate signal from the pairing strains under coculture condition. Consequently, there were no productions of the antibacterial agents from them via growth inhibition and/or stimulation.
From all 780 extracts prepared from coculture between the 40 isolated strains in our study, only 3 extracts exhibited antibacterial activity against Staphylococcus aureus ATCC25923. ese are the extracts prepared from 20-53, 53-90, and 87-90. We tried to observe the differences of secondary metabolite profile between the 2 coculture extracts   According to the strain evolutionary tree, we hypothesized that the production of the 3 extra metabolites might be a result of food source competition. ey produced antibacterial compound that can inhibit the growth of Staphylococcus aureus, in which they have closely evolutionary relationship with. e production of bioactive compound for food source competition in coculture condition might be similar to the previous report in 2017 [16]. e 3 extra metabolites should be further isolated and purified and undergo chemical structure elucidation.

Conclusion
We conducted our work with the aim of isolating and identifying high secondary metabolite production potential bacterial strains from ailand's natural resources. We used degenerate PCR genome mining together with antibacterial bioactivity analysis of the extracts prepared from their single culture and coculture condition for selection of the high potent strains. According to the PCR amplification, we found that the highest potent strain is Bacillus licheniformis RSUCC0101 whose genome has at least 6 A domains of NRPS, 9 KS domains of PKS, and 11 cyclases of TPS. e single culture extracts prepared from Bacillus safensis RSUCC0021 and Bacillus amyloliquefaciens RSUCC0282 could inhibit the growth of Staphylococcus aureus ATCC25923. Consequently, they were identified as high potential strains for production of antibacterial agents that need further investigation of the secondary metabolites in their chemical profiles. e coculture experiments results showed that Staphylococcus sp. RSUCC0020, Micrococcus luteus RSUCC0053, Staphylococcus sp. RSUCC0087, and Staphylococcus pasteuri RSUCC0090 could differently produce extra 3 secondary metabolites from their single culture extracts. e 3 metabolites could inhibit the growth of Staphylococcus aureus ATCC25923, so further chemical investigation and spectroscopic analyses of their chemical structures in future work are needed.