Pharmacological Properties, Volatile Organic Compounds, and Genome Sequences of Bacterial Endophytes from the Mangrove Plant Rhizophora apiculata Blume

Mangrove plant endophytic bacteria are prolific sources of bioactive secondary metabolites. In the present study, twenty-three endophytic bacteria were isolated from the fresh roots of the mangrove plant Rhizophora apiculata. The identification of isolates by 16S rRNA gene sequences revealed that the isolated endophytic bacteria belonged to nine genera, including Streptomyces, Bacillus, Pseudovibrio, Microbacterium, Brevibacterium, Microbulbifer, Micrococcus, Rossellomorea, and Paracoccus. The ethyl acetate extracts of the endophytic bacteria’s pharmacological properties were evaluated in vitro, including antimicrobial, antioxidant, α-amylase and α-glucosidase inhibitory, xanthine oxidase inhibitory, and cytotoxic activities. Gas chromatography–mass spectrometry (GC-MS) analyses of three high bioactive strains Bacillus sp. RAR_GA_16, Rossellomorea vietnamensis RAR_WA_32, and Bacillus sp. RAR_M1_44 identified major volatile organic compounds (VOCs) in their ethyl acetate extracts. Genome analyses identified biosynthesis gene clusters (BGCs) of secondary metabolites of the bacterial endophytes. The obtained results reveal that the endophytic bacteria from R. apiculata may be a potential source of pharmacological secondary metabolites, and further investigations of the high bioactive strains—such as fermentation and isolation of pure bioactive compounds, and heterologous expression of novel BGCs in appropriate expression hosts—may allow exploring and exploiting the promising bioactive compounds for future drug development.


Introduction
Plant endophytic bacteria are live and thrive inside plants without causing harmful effects to their host plants [1]. The endophytic bacteria provide numerous benefits to their host plants, including promoting the growth of host plants, enhancing the resistance of host plants against diseases, and increasing the tolerance of host plants to stressful environmental conditions [2][3][4]. Endophytic bacteria have been found in different parts of plants, such as roots, stems, leaves, seeds, fruits, tubers, ovules and nodules; however,

Isolation and Identification of Endophytic Bacteria
Twenty-three endophytic bacterial strains were isolated from the fresh roots of the mangrove plant Rhizophora apiculata. Based on their 16S rRNA gene sequences, the bacterial strains were identified as belonging to nine different genera, i.e., Streptomyces, Bacillus, Pseudovibrio, Microbacterium, Rossellomorea, Brevibacterium, Microbulbifer, Micrococcus, and Paracoccus ( Table 1). The phylogenic tree based on their 16S rRNA gene sequences is shown in Figure 2. All isolated strains were cultured in nutrient broth and their ethyl acetate extracts were prepared for evaluating their potential pharmacological properties.

Isolation and Identification of Endophytic Bacteria
Twenty-three endophytic bacterial strains were isolated from the fresh roots of the mangrove plant Rhizophora apiculata. Based on their 16S rRNA gene sequences, the bacterial strains were identified as belonging to nine different genera, i.e., Streptomyces, Bacillus, Pseudovibrio, Microbacterium, Rossellomorea, Brevibacterium, Microbulbifer, Micrococcus, and Paracoccus ( Table 1). The phylogenic tree based on their 16S rRNA gene sequences is shown in Figure 2. All isolated strains were cultured in nutrient broth and their ethyl acetate extracts were prepared for evaluating their potential pharmacological properties.   Bootstrap support values of branches greater than 75% are given above the corresponding branches.

Cytotoxic Activity
The cytotoxic activity of the bacterial extracts against three cancer cell lines, i.e., MCF-7 (human breast carcinoma), A549 (human lung carcinoma), and HeLa (human cervix carcinoma), are shown in Table 5. Cytotoxic bioassays showed that six extracts exhibited significant cytotoxic activity against the cell line MCF-7 with IC 50 values from 36.48 ± 2.63 to 83.24 ± 4.51 µg/mL, four extracts exhibited significant cytotoxic activity against the cell line A549 with IC 50 values from 21.52 ± 3.22 to 89.53 ± 5.31 µg/mL, and three extracts exhibited significant cytotoxic activity against the cell line A549 with IC 50 values from 41.27 ± 3.42 to 97.53 ± 5.31 µg/mL. Of these, two extracts exhibited cytotoxic activity against two cell lines, i.e., RAR_WA_32 and RAR_WA_50, and two extracts exhibited cytotoxic activity against all three tested cell lines, i.e., RAR_GA_16 and RAR_M1_44.

Genome Sequencing, Assembly, and Annotation of Biosynthesis Gene Clusters of Secondary Metabolites
In order to discover the biosynthesis gene clusters (BGCs) of secondary metabolites of the isolated bacteria, we sequenced their genomes and annotated BGCs of secondary metabolites from the genome data. The genomic features of three bacterial strains Bacillus sp. RAR_GA_16, Rossellomorea vietnamensis RAR_WA_32, and Bacillus sp. RAR_M1_44 are shown in Table 7. The size of the bacterial genomes was from 3,768,026 to 4,494,267 bp with GC contents from 40.69 to 44.09%. Assembly results showed that completion of the genomes was from 96.6 to 99.8%, with the maximum contig length from 770,470 to 2,585,281 bp and N50 contig length from 501,856 to 2,582,281 bp. Annotation of the genomes by Prokka showed that the genomes contain 3807-4610 coding sequences (CDSs), 22-27 rRNA, 81-111 tRNA, and 0-1 tmRNA. The genomes were analyzed for the presence of secondary metabolite biosynthetic gene clusters (BGCs) using antiSMASH. Annotation results showed that the genome of Bacillus sp. RAR_GA_16 contains 22 BGCs, including clusters related to the biosynthesis of linear azol(in)e-containing peptide (LAP), bacteriocin, lassopeptide, siderophore, type III polyketide (T3PKS), terpene, and several unknown compounds. Several BGCs shared their identity to the biosynthesis clusters of known compounds in Minimum Information about a Biosynthetic Gene cluster (MIBiG) database, such as paeninodin, butirosin A, butirosin B, carotenoid, thaxteramide C, and fengycin with the identities of 4-80% (Table 8).  Regarding the genome of Rossellomorea vietnamensis RAR_WA_32, annotation results showed that the genome contains 16 BGCs, including clusters related to the biosynthesis of terpene, LAP, type III polyketide, fatty acid, other unspecified ribosomally synthesised and post-translationally modified peptide product (RiPP-like), and saccharide (likely from primary metabolism). Several BGCs shared their identity to known biosynthesis clusters in the MIBiG database, such as pyxidicycline A, pyxidicycline B, A40926, and carotenoid with the identities of 3-50% (Table 8).
In the case of the genome of Bacillus sp. RAR_M1_44, it contains 16 BGCs, including clusters related to the biosynthesis of terpene, siderophore, betalactone, nonribosomal peptide (NRP), type III polyketide, bacteriocin, others and unknown compounds. Several BGCs shared their identity to the biosynthesis clusters of known compounds in the MIBiG database, such as carotenoid, fengycin, lichenysin, teichuronic acid, and bacilysin with the identities of 50-85% (Table 8). The gene structures of BCGs having the identities > 70% with known compound clusters in the MIBiG database are shown in Figure 3. The BGCs predicted from the genomes were analyzed by the Biosynthetic Genes Similarity Clustering and Prospecting Engine (BIG-SCAPE) to group the homologous BGCs into gene cluster families (GCFs), which can be responsible for the production of the same compound or similar compounds. Interestingly, no GCF was found between the BCGs from the genomes as well as between the BCGs from the genomes and the BGCs from the MIBiG database.
The ketide synthase (KS) and condensation (C) domains are the most conserved catalytic domains of polyketide synthase (PKS) and non-ribosomal peptide synthetase (NRPS) genes, respectively. Therefore, in order to predict the secondary metabolites of polyketides and non-ribosomal peptides, KS-and C-domains from the genomes were analyzed by the Natural Product Domain Seeker (NapDos). Analyses showed that the genomes of Rossellomorea vietnamensis RAR_WA_32 and Bacillus sp. RAR_M1_44 contain only KS domains without C domains (Table 9). These KS sequences shared low identities with the KS sequences of the known products such as kirromycin, mycinamicin, rifamycin, amphotericin, spinosad, and dynemicin (26-45% identity), except for fatty acids with 71-76% identity. Regarding the genome of Bacillus sp. RAR_GA_16, three KS domains and eleven C domains were found in its genome (Table 9). These sequences shared low identities with the domain sequences of the known products such as mycocerosic acid, lychenicin, surfactin, and actinomycin (22-58%), except for fatty acid (84%).  The BGCs predicted from the genomes were analyzed by the Biosynthetic Genes Similarity Clustering and Prospecting Engine (BIG-SCAPE) to group the homologous BGCs into gene cluster families (GCFs), which can be responsible for the production of the same compound or similar compounds. Interestingly, no GCF was found between the BCGs from the genomes as well as between the BCGs from the genomes and the BGCs from the MIBiG database.
The ketide synthase (KS) and condensation (C) domains are the most conserved catalytic domains of polyketide synthase (PKS) and non-ribosomal peptide synthetase (NRPS) genes, respectively. Therefore, in order to predict the secondary metabolites of polyketides and non-ribosomal peptides, KS-and C-domains from the genomes were analyzed by the Natural Product Domain Seeker (NapDos). Analyses showed that the genomes of Rossellomorea vietnamensis RAR_WA_32 and Bacillus sp. RAR_M1_44 contain only KS domains without C domains (Table 9). These KS sequences shared low identities with the KS sequences of the known products such as kirromycin, mycinamicin, rifamycin, amphotericin, spinosad, and dynemicin (26-45% identity), except for fatty acids with 71-76% identity. Regarding the genome of Bacillus sp. RAR_GA_16, three KS domains and eleven C domains were found in its genome (Table 9). These sequences shared low identities with the domain sequences of the known products such as mycocerosic acid, lychenicin, surfactin, and actinomycin (22-58%), except for fatty acid (84%).

Discussion
Mangrove plants harbour diverse endophytic bacteria and fungi. Reviews on mangrove plant-associated microbial secondary metabolites revealed that the secondary metabolites possess promising pharmacological properties [8][9][10][11]. Unsurprisingly, mangrove endophytic microbes have been attracting the considerable attention of pharmacological investigators because of their bioactive secondary metabolites, especially when their host plants are also used as native traditional/folk medicines [69]. Over the last decade, almost 1000 new natural products have been reported from mangrove associated microbes, of these,~850 are derived from fungi and~120 are derived from bacteria (the majority of them derived from endophytes) [10].
The ethyl acetate extracts of three endophytic bacteria with the most potential biological properties, i.e., Bacillus sp. RAR_GA_16, Rossellomorea vietnamensis RAR_WA_32, and Bacillus sp. RAR_M1_44, were investigated for their volatile chemical components by GC-MS analyses. Interestingly, many compounds identified in the extracts of three endophytic bacteria have been reported to have biological activities (e.g., antibacterial, antifungal, antivirus, anticancer, anti-inflammatory, antioxidant, nematicidal, anti-quorum sensing, tyrosinase inhibitory, antibiofilm, and anti-mutagenic activities) by previous studies (Table 6). These GC-MS analyses support the biological activity evaluation results of the ethyl acetate extracts. Additionally, it is noted that the identified VOCs from the extracts only accounted for 44.05%, 80.98%, and 4.10% of the total amount of VOCs in the extracts of Bacillus sp. RAR_GA_16, Rossellomorea vietnamensis RAR_WA_32, and Bacillus sp. RAR_M1_44, respectively, whereas many VOCs in the extracts were not identified by their low match quality with the spectra data of known compounds (Supplementary Tables S1-S3). This implies that many VOCs in the extracts may be undescribed or novel compounds.
In the present study, the genomes of three potential endophytic bacteria, i.e., Bacillus sp. RAR_GA_16, Rossellomorea vietnamensis RAR_WA_32, and Bacillus sp. RAR_M1_44, were also sequenced and the presence of BGCs of secondary metabolites in their genomes was discovered. Annotations using the antiSMASH found 16-22 BGCs, 3-5 KS domains, and 11 C domains in the bacterial genomes. However, the majority of BGCs, KS-and C-domains shared low identity with BGC, KS-and C-sequences in the known product database. Additionally, the BIG-SCAPE analysis also indicated that no BGCs predicted from the genomes were grouped in GCFs of known products in the MIBiG database. These findings suggest that the BGCs of the endophytic bacteria may biosynthesize novel compounds or compounds for which their BGCs have not been known. Interestingly, no GCF was found between the BGCs predicted from the genomes, implying that the BGCs of the endophytic bacteria may biosynthesize dissimilar compounds. The obtained results in the present study indicate that these endophytic bacteria may be potential sources for the discovery of novel bioactive metabolites, and further investigations, e.g., fermentation and isolation of pure bioactive compounds, as well as heterologous expression of novel BGCs in appropriate expression hosts, will allow exploring and exploiting novel bioactive compounds as well as demonstrating their biosynthesis pathways.

Plant Collection
The mangrove plant Rhizophora apiculata (Figure 1

Isolation and Identification of the Endophytic Bacteria
The fresh plant root systems were washed thoroughly with running water to remove soil particles. The root surfaces were then washed in sterile distilled water six times, soaked in 70% ethanol for 5 min followed by 5% sodium hypochlorite for 10 min, and again in 70% ethanol for 60 s before washing in distilled water three times.
The representative isolates with different morphotypes were identified by colony PCR [85]. In brief, for cell lysis, the culture broths (2 mL) were centrifuged at 14,000× g for 10 min and the obtained pellets were suspended in 50 µL nuclease-free water. Subsequently, the cell suspension was stored at −20 • C for 2 h, followed by incubation at 98 • C for 10 min. The 16S rRNA gene of the isolates was directly amplified with universal primers 27f (5 -AGAGTTTGATCCTGGCTCAG-3 ) and 1492r (5 -GGTTACCTTGTTACGACTT-3 ) [86] with the following PCR program: an initial denaturation at 94 • C for 5 min, followed by 30 cycles of denaturation at 94 • C for 1 min, annealing at 56 • C for 50 s, amplification at 72 • C for 1.5 min, and a final extension at 72 • C for 7 min. The 16S rRNA gene sequencing was carried by the ABI PRISM 3100 ® Genetic Analyzer (Applied Bioscience and Hitachi, Foster City, CA, USA). The sequences were quality checked, and low-quality regions were removed from the sequence ends using BioEdit software v.7.2.6.1. The quality-checked sequences of the isolates were compared to available sequences in the NCBI GenBank using the BLAST searching program with the megablast algorithm and the database nt. The sequences were aligned using the ClustalW algorithm and the phylogenetic tree of 16S rRNA sequences was created by the neighbor joining algorithm with 1000 bootstraps using MEGA v.7.0.0.

Preparation of Ethyl Acetate Extracts from the Culture Broths
The endophytic bacterial strains were cultured in 500 mL nutrient broth (NB, Himedia, Mumbai, India) for 7 days at 37 • C under the shaking condition at 150 rpm, and the cultures were then centrifuged at 10.000 rpm for 10 min. The cell-free supernatants were extracted with ethyl acetate (1:1 v/v, 5 times) overnight at room temperature, and the ethyl acetate extractions were then evaporated under the reduced pressure for 12-24 h at 50 • C) to remove ethyl acetate and obtain the crude extracts.

Antimicrobial Activity
Antimicrobial activity of the bacterial extracts was tested against five reference microbes obtained from Mientrung Institute for Scientific Research, Thua Thien Hue province, Vietnam, i.e., Staphylococcus aureus ATCC 25923, Enterococcus faecalis ATCC 29212, Escherichia coli ATCC 25922, Pseudomonas aegurinosa ATCC 27853, and Candida albicans ATCC 10231.
Minimum inhibitory concentrations (MICs) of the extracts against the reference microorganisms were determined using the broth microdilution method as described by Dat et al. [85]. Briefly, 100 µL of the bacterial inoculum (1 × 10 6 CFU/mL) was added to wells containing 100 µL of the extracts at a range of different concentrations in 96-well plates. The plate was incubated at 37 • C for 24 h, the absorbance at 630 nm was then measured using ELx800 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA). MICs of the antibacterial extracts were determined as the lowest concentration, at which there was no growth of the bacteria. For the yeast, 100 µL of the inoculum (2 to 5 × 10 5 CFU/mL for yeast) was added to wells containing 100 µL of the extracts at a range of different concentrations in 96-well plates. The plate was incubated at 28 • C for 48 h. MICs of the anti-yeast extracts were determined as the lowest concentration, at which there was no growth of the yeast by the absorbance at 530 nm using an ELx800 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA). The antibiotics ciprofloxacin and fluconazole were used as positive controls for the tested bacteria and yeast, respectively.

Antioxidant Activity
The antioxidant effect of the extracts was determined by DPPH and ABTS radical scavenging assays [70].
DPPH radical scavenging effect of the extracts was determined by measuring the decrease in absorbance of DPPH radical solution in the presence of the extracts. In brief, 10 µL of extracts was added to 190 µL of DPPH (0.1 mg/mL) in 96-well plates. The solution was mixed for 1 min and incubated at room temperature for 30 min, and the absorbance of the reaction mixture was then recorded at 517 nm using an ELx800 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA). Ascorbic acid was used as a positive control. The DPPH radical scavenging activity was calculated as follows: DPPH scavenging activity (%) = 100 × [Ac − (As − Asb)/Ac]. Where: Ac is the absorbance of the control (only DPPH solution), As is absorbance of sample (extract with DPPH), and Asb is the absorbance of the sample blank (extract without DPPH).
ABTS radical scavenging effect of the extracts was determined by measuring the decrease in absorbance of ABTS radical solution in the presence of the extracts. In brief, two solutions (ABTS 7 mM and potassium persulfate 2.45 mM) were mixed and allowed to stand in the dark at room temperature for 16 h before use in order to produce ABTS radical solution. The ABTS radical solution was then diluted with ethanol to give an absorbance of 0.700 ± 0.02 at 734 nm. Ten microliters of the extracts was added to 190 µL of ABTS radical solution in 96-well plates. The mixture was incubated at room temperature for 10 min, and the absorbance of the reaction was then recorded at 734 nm using ELx800 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA). Ascorbic acid was used as a positive control. The ABTS radical scavenging activity was calculated as follows: ABTS scavenging activity (%) = 100 × [Ac − (As − Asb)/Ac]. Where: Ac is the absorbance of the control (only ABTS solution), As is absorbance of sample (extract with ABTS), and Asb is the absorbance of the sample blank (extract without ABTS).

α-Amylase and α-Glucosidase Activities
The α-amylase (A8220, Sigma-Aldrich, St. Louis, MO, USA) enzyme inhibitory effect of the extracts was determined according to the described method by Dat et al. [70]. In brief, starch azure was suspended in 0.05 M Tris-HCl buffer (pH 6.9) containing 0.01 M CaCl 2 , and the substrate solution was then boiled for 5 min and pre-incubated at 37 • C for 5 min. The reaction mixture consisting of 50 µL of the extract was incubated with 50 µL of the substrate solution and 25 µL of α-amylase solution in Tris-HCl buffer (2 U/mL) in 96-well plates at 37 • C for 10 min. The reaction was stopped by adding 75 µL of acetic acid 50%, and the reaction solution was then centrifuged at 3000 rpm for 5 min at 4 • C. The absorbance of the supernatant was recorded at 650 nm using an ELx800 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA). The inhibition activity was calculated as follows: inhibition (%) = 100 × [1 − (As − Abs)/(Ac − Acb)]. Where: As is the absorbance of the sample (extract with enzyme), Asb is the absorbance of the sample blank (extract without enzyme), Ac is the absorbance of the control (100% enzyme activity, only solvent with enzyme), and Acb is the absorbance of the control blank (0% enzyme activity, only solvent without enzyme). Acarbose was used as a positive control.
α-Glucosidase (G0660, Sigma-Aldrich, St. Louis, MO, USA) enzyme inhibitory effect of the extracts was determined according to the described method by . In brief, the reaction mixture consisting of 50 µL of the extract was incubated with 100 µL of 0.1 M potassium phosphate buffer (pH 6.8) containing α-glucosidase solution (0.5 U/mL) in 96-well plates at 37 • C for 10 min. The reaction was started by adding 50 µL of 5 mM 4-Nitrophenyl β-D-glucopyranoside (pNPG), followed by incubation at 37 • C for 30 min. The absorbance of released p-nitrophenol was recorded at 405 nm using an ELx800 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA). The inhibition activity was calculated as follows: inhibition (%) = 100 × [1 − (As − Asb)/(Ac − Acb)]. Where: As is the absorbance of the sample (extract with enzyme), and Asb is the absorbance of the sample blank (extract without enzyme), Ac is the absorbance of the control (100% enzyme activity, only solvent with enzyme), and Acb is the absorbance of control blank (0% enzyme activity, only solvent without enzyme). Acarbose was used as a positive control.

Xanthine Oxidase Inhibitory Activity
The xanthine oxidase (XO) inhibitory effect of the extracts was determined according to the method described by Nguyen et al. [87] with minor modifications. In brief, the reaction mixture consisting of 50 µL of the extract, 35 µL of 70 mM phosphate buffer (pH7.5), and 30 µL of enzyme solution (0.01 units/mL in 70 mM phosphate buffer, pH 7.5) was prepared immediately before use. Subsequently, the reaction mixture was preincubated at 25 • C for 15 min, and the reaction was then initiated by adding 60 µL of substrate solution (150 mM xanthine in 70 mM phosphate buffer, pH 7.5). The reaction mixture was incubated at 25 • C for 30 min, and the reaction was then stopped by adding 25 µL of 1 N HCl. The absorbance of the mixture was measured at 290 nm using an ELx800 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA). The XO inhibition activity was calculated as follows: inhibition (%) = 100 × [1 − (As − Asb)/(Ac − Acb)]. Where: As is the absorbance of the sample (extract with enzyme), Asb is the absorbance of the sample blank (extract without enzyme), Ac is the absorbance of the control (100% enzyme activity, only solvent with enzyme), and Acb is the absorbance of the control blank (0% enzyme activity, only solvent without enzyme). Allopurinol was used as a positive control.

Cytotoxic Activity
Cytotoxic activity of the extracts against three cancer cell lines A549 (human lung carcinoma), MCF-7 (human breast carcinoma), and HeLa (human cervix carcinoma) was determined by Sulforhodamine B (SRB) assay as previously described by Skehan et al. [88]. Camptothecin was used as a positive control.

Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
GC-MS analyses were conducted using the Agilent 7890B gas chromatograph-assisted Agilent 5977A mass detector (Agilent Technologies, Stanta Clara, CA, USA). An HP-5MS capillary column (30 m × 0.250 mm × 0.25 µM film thickness; Agilent Technologies, Stanta Clara, CA, USA) was used for separation. The samples were diluted in hexane (1:10) and volumes of 1 µL were injected (splitless mode) into the GC system at a flow rate of 1 mL/min. Helium (99.999% of purity) was the carrier gas with a flow rate of 1 mL/min. The oven temperature was set at 60 • C and held for 2 min, then increased to 260 • C at a rate of 5 • C/min and held at this temperature for 1 min. The inlet temperature was 260 • C and the ionization source temperature was 280 • C. The solvent delay was 3.00 min. The MS detector was operated in the EI mode at 70 eV, in the range of m/z 50-550, full scan mode. Data handling was performed using the Agilent ChemStation software C.01.10 (Agilent Technologies, Stanta Clara, CA, USA). The compounds were identified by comparing the spectra with a stored MS library (W8N08 and NIST08) with minimum matching quality of 90%. The relative percent of individual components was calculated based on GC peak areas.

Genome Sequencing, Assembly and Annotation of Biosynthesis Gene Clusters of Secondary Metabolites
The endophytic bacterial strains were cultured in nutrient broth (NB, Himedia, Mumbai, India) overnight at 37 • C under the shaking condition at 150 rpm. Cells were harvested by centrifugation at 10,000× g for 10 min, and genomic DNA was isolated using the QIAamp DNA Mini Kit according to the manufacturer's protocol (QIAGEN, Hilden, Germany). The genome of the bacterial strains was sequenced by PacBio Sequel technology (PacBio, Menlo Park, CA, USA) according to the manufacturer's instructions.

Conclusions
In the present study, twenty-three endophytic bacteria belonging to nine genera Streptomyces, Bacillus, Pseudovibrio, Microbacterium, Brevibacterium, Microbulbifer, Micrococcus, Rossellomorea, and Paracoccus were isolated from the fresh roots of the mangrove plant R. apiculata. The ethyl acetate extracts of the endophytic bacteria were evaluated for their pharmacological properties including antimicrobial, antioxidant, α-amylase and α-glucosidase inhibitory, xanthine oxidase inhibitory and cytotoxic activities. GC-MS analyses identified major volatile organic compounds from the ethyl acetate extract of the most potential strains Bacillus sp. RAR_GA_16, Rossellomorea vietnamensis RAR_WA_32, and Bacillus sp. RAR_M1_44. Genome studies found gene clusters related to the biosynthesis of secondary metabolites from the bacteria endophytes. Further investigations of the high bioactive strains-such as fermentation and isolation of pure bioactive compounds, and heterologous expression of novel BGCs in appropriate expression hosts-may allow exploring and exploiting the promising bioactive compounds for future drug development.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/antibiotics10121491/s1, Table S1. Comparison of the GC-MS peaks of the extract of Bacillus sp. RAR_GA_16 with the spectra library. Table S2. Comparison of the GC-MS peaks of the extract of R. vietnamensis RAR_WA_32 with the spectra library. Table S3. Comparison of the GC-MS peaks of the extract of Bacillus sp. RAR_M1_44 with the spectra library.

Conflicts of Interest:
The authors declare no conflict of interest.